From 04600794958f1833f5571c6cde40f260ab557f55 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 5 Aug 2010 17:45:15 +0200 Subject: cfg80211: support sysfs namespaces Enable using network namespaces with wireless devices even when sysfs is enabled using the same infrastructure that was built for netdevs. Signed-off-by: Johannes Berg Acked-by: "Eric W. Biederman" Signed-off-by: John W. Linville --- net/core/net-sysfs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net/core') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index af4dfbadf2a0..7d748542d97e 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -789,12 +789,13 @@ static const void *net_netlink_ns(struct sock *sk) return sock_net(sk); } -static struct kobj_ns_type_operations net_ns_type_operations = { +struct kobj_ns_type_operations net_ns_type_operations = { .type = KOBJ_NS_TYPE_NET, .current_ns = net_current_ns, .netlink_ns = net_netlink_ns, .initial_ns = net_initial_ns, }; +EXPORT_SYMBOL_GPL(net_ns_type_operations); static void net_kobj_ns_exit(struct net *net) { -- cgit v1.2.3 From bfb564e7391340638afe4ad67744a8f3858e7566 Mon Sep 17 00:00:00 2001 From: Krishna Kumar Date: Wed, 4 Aug 2010 06:15:52 +0000 Subject: core: Factor out flow calculation from get_rps_cpu Factor out flow calculation code from get_rps_cpu, since other functions can use the same code. Revisions: v2 (Ben): Separate flow calcuation out and use in select queue. v3 (Arnd): Don't re-implement MIN. v4 (Changli): skb->data points to ethernet header in macvtap, and make a fast path. Tested macvtap with this patch. v5 (Changli): - Cache skb->rxhash in skb_get_rxhash - macvtap may not have pow(2) queues, so change code for queue selection. (Arnd): - Use first available queue if all fails. Signed-off-by: Krishna Kumar Signed-off-by: David S. Miller --- include/linux/skbuff.h | 9 +++++ net/core/dev.c | 106 +++++++++++++++++++++++++++++-------------------- 2 files changed, 71 insertions(+), 44 deletions(-) (limited to 'net/core') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 77eb60d2b496..d8050382b189 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -558,6 +558,15 @@ extern unsigned int skb_find_text(struct sk_buff *skb, unsigned int from, unsigned int to, struct ts_config *config, struct ts_state *state); +extern __u32 __skb_get_rxhash(struct sk_buff *skb); +static inline __u32 skb_get_rxhash(struct sk_buff *skb) +{ + if (!skb->rxhash) + skb->rxhash = __skb_get_rxhash(skb); + + return skb->rxhash; +} + #ifdef NET_SKBUFF_DATA_USES_OFFSET static inline unsigned char *skb_end_pointer(const struct sk_buff *skb) { diff --git a/net/core/dev.c b/net/core/dev.c index 1ae654391442..586a11cb4398 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2259,69 +2259,41 @@ static inline void ____napi_schedule(struct softnet_data *sd, __raise_softirq_irqoff(NET_RX_SOFTIRQ); } -#ifdef CONFIG_RPS - -/* One global table that all flow-based protocols share. */ -struct rps_sock_flow_table *rps_sock_flow_table __read_mostly; -EXPORT_SYMBOL(rps_sock_flow_table); - /* - * get_rps_cpu is called from netif_receive_skb and returns the target - * CPU from the RPS map of the receiving queue for a given skb. - * rcu_read_lock must be held on entry. + * __skb_get_rxhash: calculate a flow hash based on src/dst addresses + * and src/dst port numbers. Returns a non-zero hash number on success + * and 0 on failure. */ -static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, - struct rps_dev_flow **rflowp) +__u32 __skb_get_rxhash(struct sk_buff *skb) { + int nhoff, hash = 0; struct ipv6hdr *ip6; struct iphdr *ip; - struct netdev_rx_queue *rxqueue; - struct rps_map *map; - struct rps_dev_flow_table *flow_table; - struct rps_sock_flow_table *sock_flow_table; - int cpu = -1; u8 ip_proto; - u16 tcpu; u32 addr1, addr2, ihl; union { u32 v32; u16 v16[2]; } ports; - if (skb_rx_queue_recorded(skb)) { - u16 index = skb_get_rx_queue(skb); - if (unlikely(index >= dev->num_rx_queues)) { - WARN_ONCE(dev->num_rx_queues > 1, "%s received packet " - "on queue %u, but number of RX queues is %u\n", - dev->name, index, dev->num_rx_queues); - goto done; - } - rxqueue = dev->_rx + index; - } else - rxqueue = dev->_rx; - - if (!rxqueue->rps_map && !rxqueue->rps_flow_table) - goto done; - - if (skb->rxhash) - goto got_hash; /* Skip hash computation on packet header */ + nhoff = skb_network_offset(skb); switch (skb->protocol) { case __constant_htons(ETH_P_IP): - if (!pskb_may_pull(skb, sizeof(*ip))) + if (!pskb_may_pull(skb, sizeof(*ip) + nhoff)) goto done; - ip = (struct iphdr *) skb->data; + ip = (struct iphdr *) skb->data + nhoff; ip_proto = ip->protocol; addr1 = (__force u32) ip->saddr; addr2 = (__force u32) ip->daddr; ihl = ip->ihl; break; case __constant_htons(ETH_P_IPV6): - if (!pskb_may_pull(skb, sizeof(*ip6))) + if (!pskb_may_pull(skb, sizeof(*ip6) + nhoff)) goto done; - ip6 = (struct ipv6hdr *) skb->data; + ip6 = (struct ipv6hdr *) skb->data + nhoff; ip_proto = ip6->nexthdr; addr1 = (__force u32) ip6->saddr.s6_addr32[3]; addr2 = (__force u32) ip6->daddr.s6_addr32[3]; @@ -2330,6 +2302,7 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, default: goto done; } + switch (ip_proto) { case IPPROTO_TCP: case IPPROTO_UDP: @@ -2338,8 +2311,9 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, case IPPROTO_AH: case IPPROTO_SCTP: case IPPROTO_UDPLITE: - if (pskb_may_pull(skb, (ihl * 4) + 4)) { - ports.v32 = * (__force u32 *) (skb->data + (ihl * 4)); + if (pskb_may_pull(skb, (ihl * 4) + 4 + nhoff)) { + ports.v32 = * (__force u32 *) (skb->data + nhoff + + (ihl * 4)); if (ports.v16[1] < ports.v16[0]) swap(ports.v16[0], ports.v16[1]); break; @@ -2352,11 +2326,55 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, /* get a consistent hash (same value on both flow directions) */ if (addr2 < addr1) swap(addr1, addr2); - skb->rxhash = jhash_3words(addr1, addr2, ports.v32, hashrnd); - if (!skb->rxhash) - skb->rxhash = 1; -got_hash: + hash = jhash_3words(addr1, addr2, ports.v32, hashrnd); + if (!hash) + hash = 1; + +done: + return hash; +} +EXPORT_SYMBOL(__skb_get_rxhash); + +#ifdef CONFIG_RPS + +/* One global table that all flow-based protocols share. */ +struct rps_sock_flow_table *rps_sock_flow_table __read_mostly; +EXPORT_SYMBOL(rps_sock_flow_table); + +/* + * get_rps_cpu is called from netif_receive_skb and returns the target + * CPU from the RPS map of the receiving queue for a given skb. + * rcu_read_lock must be held on entry. + */ +static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, + struct rps_dev_flow **rflowp) +{ + struct netdev_rx_queue *rxqueue; + struct rps_map *map; + struct rps_dev_flow_table *flow_table; + struct rps_sock_flow_table *sock_flow_table; + int cpu = -1; + u16 tcpu; + + if (skb_rx_queue_recorded(skb)) { + u16 index = skb_get_rx_queue(skb); + if (unlikely(index >= dev->num_rx_queues)) { + WARN_ONCE(dev->num_rx_queues > 1, "%s received packet " + "on queue %u, but number of RX queues is %u\n", + dev->name, index, dev->num_rx_queues); + goto done; + } + rxqueue = dev->_rx + index; + } else + rxqueue = dev->_rx; + + if (!rxqueue->rps_map && !rxqueue->rps_flow_table) + goto done; + + if (!skb_get_rxhash(skb)) + goto done; + flow_table = rcu_dereference(rxqueue->rps_flow_table); sock_flow_table = rcu_dereference(rps_sock_flow_table); if (flow_table && sock_flow_table) { -- cgit v1.2.3 From 01414802054c382072b6cb9a1bdc6e243c74b2d5 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Tue, 17 Aug 2010 02:31:15 -0700 Subject: ethtool: Provide a default implementation of ethtool_ops::get_drvinfo The driver name and bus address for a net_device can normally be found through the driver model now. Instead of requiring drivers to provide this information redundantly through the ethtool_ops::get_drvinfo operation, use the driver model to do so if the driver does not define the operation. Since ETHTOOL_GDRVINFO no longer requires the driver to implement any operations, do not require net_device::ethtool_ops to be set either. Remove implementations of get_drvinfo and ethtool_ops that provide only this information. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- drivers/firewire/net.c | 13 ------------- drivers/ieee1394/eth1394.c | 16 ---------------- drivers/net/bmac.c | 7 ------- drivers/net/fec_mpc52xx.c | 6 ------ drivers/net/pasemi_mac_ethtool.c | 16 ---------------- drivers/net/pcmcia/3c574_cs.c | 13 ------------- drivers/net/pcmcia/axnet_cs.c | 13 ------------- drivers/net/pcmcia/ibmtr_cs.c | 13 ------------- drivers/net/pcmcia/pcnet_cs.c | 16 ---------------- drivers/net/sc92031.c | 11 ----------- drivers/net/tulip/xircom_cb.c | 15 --------------- drivers/net/usb/hso.c | 9 --------- drivers/net/usb/kaweth.c | 9 --------- drivers/net/virtio_net.c | 14 -------------- drivers/net/wireless/ray_cs.c | 16 ---------------- drivers/net/wireless/wl3501_cs.c | 11 ----------- net/core/ethtool.c | 33 +++++++++++++++++++++++---------- 17 files changed, 23 insertions(+), 208 deletions(-) (limited to 'net/core') diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c index da17d409a244..51e8a35ebd7e 100644 --- a/drivers/firewire/net.c +++ b/drivers/firewire/net.c @@ -8,7 +8,6 @@ #include #include -#include #include #include #include @@ -1359,17 +1358,6 @@ static int fwnet_change_mtu(struct net_device *net, int new_mtu) return 0; } -static void fwnet_get_drvinfo(struct net_device *net, - struct ethtool_drvinfo *info) -{ - strcpy(info->driver, KBUILD_MODNAME); - strcpy(info->bus_info, "ieee1394"); -} - -static const struct ethtool_ops fwnet_ethtool_ops = { - .get_drvinfo = fwnet_get_drvinfo, -}; - static const struct net_device_ops fwnet_netdev_ops = { .ndo_open = fwnet_open, .ndo_stop = fwnet_stop, @@ -1388,7 +1376,6 @@ static void fwnet_init_dev(struct net_device *net) net->hard_header_len = FWNET_HLEN; net->type = ARPHRD_IEEE1394; net->tx_queue_len = 10; - SET_ETHTOOL_OPS(net, &fwnet_ethtool_ops); } /* caller must hold fwnet_device_mutex */ diff --git a/drivers/ieee1394/eth1394.c b/drivers/ieee1394/eth1394.c index bc289e367e30..63403822330e 100644 --- a/drivers/ieee1394/eth1394.c +++ b/drivers/ieee1394/eth1394.c @@ -58,7 +58,6 @@ #include #include #include -#include #include #include #include @@ -173,8 +172,6 @@ static netdev_tx_t ether1394_tx(struct sk_buff *skb, struct net_device *dev); static void ether1394_iso(struct hpsb_iso *iso); -static const struct ethtool_ops ethtool_ops; - static int ether1394_write(struct hpsb_host *host, int srcid, int destid, quadlet_t *data, u64 addr, size_t len, u16 flags); static void ether1394_add_host(struct hpsb_host *host); @@ -525,8 +522,6 @@ static void ether1394_init_dev(struct net_device *dev) dev->header_ops = ðer1394_header_ops; dev->netdev_ops = ðer1394_netdev_ops; - SET_ETHTOOL_OPS(dev, ðtool_ops); - dev->watchdog_timeo = ETHER1394_TIMEOUT; dev->flags = IFF_BROADCAST | IFF_MULTICAST; dev->features = NETIF_F_HIGHDMA; @@ -1695,17 +1690,6 @@ fail: return NETDEV_TX_OK; } -static void ether1394_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *info) -{ - strcpy(info->driver, driver_name); - strcpy(info->bus_info, "ieee1394"); /* FIXME provide more detail? */ -} - -static const struct ethtool_ops ethtool_ops = { - .get_drvinfo = ether1394_get_drvinfo -}; - static int __init ether1394_init_module(void) { int err; diff --git a/drivers/net/bmac.c b/drivers/net/bmac.c index 959add2410bf..9322699bb31c 100644 --- a/drivers/net/bmac.c +++ b/drivers/net/bmac.c @@ -1233,15 +1233,8 @@ static void bmac_reset_and_enable(struct net_device *dev) } spin_unlock_irqrestore(&bp->lock, flags); } -static void bmac_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) -{ - struct bmac_data *bp = netdev_priv(dev); - strcpy(info->driver, "bmac"); - strcpy(info->bus_info, dev_name(&bp->mdev->ofdev.dev)); -} static const struct ethtool_ops bmac_ethtool_ops = { - .get_drvinfo = bmac_get_drvinfo, .get_link = ethtool_op_get_link, }; diff --git a/drivers/net/fec_mpc52xx.c b/drivers/net/fec_mpc52xx.c index e3e10b4add9c..e9f5d030bc26 100644 --- a/drivers/net/fec_mpc52xx.c +++ b/drivers/net/fec_mpc52xx.c @@ -771,11 +771,6 @@ static void mpc52xx_fec_reset(struct net_device *dev) /* ethtool interface */ -static void mpc52xx_fec_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *info) -{ - strcpy(info->driver, DRIVER_NAME); -} static int mpc52xx_fec_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { @@ -810,7 +805,6 @@ static void mpc52xx_fec_set_msglevel(struct net_device *dev, u32 level) } static const struct ethtool_ops mpc52xx_fec_ethtool_ops = { - .get_drvinfo = mpc52xx_fec_get_drvinfo, .get_settings = mpc52xx_fec_get_settings, .set_settings = mpc52xx_fec_set_settings, .get_link = ethtool_op_get_link, diff --git a/drivers/net/pasemi_mac_ethtool.c b/drivers/net/pasemi_mac_ethtool.c index fefa79e34b95..4825959a0efe 100644 --- a/drivers/net/pasemi_mac_ethtool.c +++ b/drivers/net/pasemi_mac_ethtool.c @@ -90,21 +90,6 @@ pasemi_mac_ethtool_set_settings(struct net_device *netdev, return phy_ethtool_sset(phydev, cmd); } -static void -pasemi_mac_ethtool_get_drvinfo(struct net_device *netdev, - struct ethtool_drvinfo *drvinfo) -{ - struct pasemi_mac *mac; - mac = netdev_priv(netdev); - - /* clear and fill out info */ - memset(drvinfo, 0, sizeof(struct ethtool_drvinfo)); - strncpy(drvinfo->driver, "pasemi_mac", 12); - strcpy(drvinfo->version, "N/A"); - strcpy(drvinfo->fw_version, "N/A"); - strncpy(drvinfo->bus_info, pci_name(mac->pdev), 32); -} - static u32 pasemi_mac_ethtool_get_msglevel(struct net_device *netdev) { @@ -164,7 +149,6 @@ static void pasemi_mac_get_strings(struct net_device *netdev, u32 stringset, const struct ethtool_ops pasemi_mac_ethtool_ops = { .get_settings = pasemi_mac_ethtool_get_settings, .set_settings = pasemi_mac_ethtool_set_settings, - .get_drvinfo = pasemi_mac_ethtool_get_drvinfo, .get_msglevel = pasemi_mac_ethtool_get_msglevel, .set_msglevel = pasemi_mac_ethtool_set_msglevel, .get_link = ethtool_op_get_link, diff --git a/drivers/net/pcmcia/3c574_cs.c b/drivers/net/pcmcia/3c574_cs.c index c683f77c6f42..9a1840b67e78 100644 --- a/drivers/net/pcmcia/3c574_cs.c +++ b/drivers/net/pcmcia/3c574_cs.c @@ -83,7 +83,6 @@ earlier 3Com products. #include #include #include -#include #include #include @@ -238,7 +237,6 @@ static int el3_rx(struct net_device *dev, int worklimit); static int el3_close(struct net_device *dev); static void el3_tx_timeout(struct net_device *dev); static int el3_ioctl(struct net_device *dev, struct ifreq *rq, int cmd); -static const struct ethtool_ops netdev_ethtool_ops; static void set_rx_mode(struct net_device *dev); static void set_multicast_list(struct net_device *dev); @@ -285,7 +283,6 @@ static int tc574_probe(struct pcmcia_device *link) link->conf.ConfigIndex = 1; dev->netdev_ops = &el3_netdev_ops; - SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops); dev->watchdog_timeo = TX_TIMEOUT; return tc574_config(link); @@ -1065,16 +1062,6 @@ static int el3_rx(struct net_device *dev, int worklimit) return worklimit; } -static void netdev_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *info) -{ - strcpy(info->driver, "3c574_cs"); -} - -static const struct ethtool_ops netdev_ethtool_ops = { - .get_drvinfo = netdev_get_drvinfo, -}; - /* Provide ioctl() calls to examine the MII xcvr state. */ static int el3_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { diff --git a/drivers/net/pcmcia/axnet_cs.c b/drivers/net/pcmcia/axnet_cs.c index 5f05ffb240cc..a6e37b29e3bf 100644 --- a/drivers/net/pcmcia/axnet_cs.c +++ b/drivers/net/pcmcia/axnet_cs.c @@ -32,7 +32,6 @@ #include #include #include -#include #include #include #include @@ -86,7 +85,6 @@ static netdev_tx_t axnet_start_xmit(struct sk_buff *skb, static struct net_device_stats *get_stats(struct net_device *dev); static void set_multicast_list(struct net_device *dev); static void axnet_tx_timeout(struct net_device *dev); -static const struct ethtool_ops netdev_ethtool_ops; static irqreturn_t ei_irq_wrapper(int irq, void *dev_id); static void ei_watchdog(u_long arg); static void axnet_reset_8390(struct net_device *dev); @@ -171,7 +169,6 @@ static int axnet_probe(struct pcmcia_device *link) dev->netdev_ops = &axnet_netdev_ops; - SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops); dev->watchdog_timeo = TX_TIMEOUT; return axnet_config(link); @@ -658,16 +655,6 @@ reschedule: add_timer(&info->watchdog); } -static void netdev_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *info) -{ - strcpy(info->driver, "axnet_cs"); -} - -static const struct ethtool_ops netdev_ethtool_ops = { - .get_drvinfo = netdev_get_drvinfo, -}; - /*====================================================================*/ static int axnet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) diff --git a/drivers/net/pcmcia/ibmtr_cs.c b/drivers/net/pcmcia/ibmtr_cs.c index b0d06a3d962f..cbcda123b1bf 100644 --- a/drivers/net/pcmcia/ibmtr_cs.c +++ b/drivers/net/pcmcia/ibmtr_cs.c @@ -52,7 +52,6 @@ #include #include #include -#include #include #include #include @@ -107,16 +106,6 @@ typedef struct ibmtr_dev_t { struct tok_info *ti; } ibmtr_dev_t; -static void netdev_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *info) -{ - strcpy(info->driver, "ibmtr_cs"); -} - -static const struct ethtool_ops netdev_ethtool_ops = { - .get_drvinfo = netdev_get_drvinfo, -}; - static irqreturn_t ibmtr_interrupt(int irq, void *dev_id) { ibmtr_dev_t *info = dev_id; struct net_device *dev = info->dev; @@ -159,8 +148,6 @@ static int __devinit ibmtr_attach(struct pcmcia_device *link) info->dev = dev; - SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops); - return ibmtr_config(link); } /* ibmtr_attach */ diff --git a/drivers/net/pcmcia/pcnet_cs.c b/drivers/net/pcmcia/pcnet_cs.c index c3edfe4c2651..2e1348c0033e 100644 --- a/drivers/net/pcmcia/pcnet_cs.c +++ b/drivers/net/pcmcia/pcnet_cs.c @@ -35,7 +35,6 @@ #include #include #include -#include #include #include #include @@ -100,7 +99,6 @@ static void pcnet_release(struct pcmcia_device *link); static int pcnet_open(struct net_device *dev); static int pcnet_close(struct net_device *dev); static int ei_ioctl(struct net_device *dev, struct ifreq *rq, int cmd); -static const struct ethtool_ops netdev_ethtool_ops; static irqreturn_t ei_irq_wrapper(int irq, void *dev_id); static void ei_watchdog(u_long arg); static void pcnet_reset_8390(struct net_device *dev); @@ -628,8 +626,6 @@ static int pcnet_config(struct pcmcia_device *link) ei_status.word16 = 1; ei_status.reset_8390 = &pcnet_reset_8390; - SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops); - if (info->flags & (IS_DL10019|IS_DL10022)) mii_phy_probe(dev); @@ -1143,18 +1139,6 @@ reschedule: /*====================================================================*/ -static void netdev_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *info) -{ - strcpy(info->driver, "pcnet_cs"); -} - -static const struct ethtool_ops netdev_ethtool_ops = { - .get_drvinfo = netdev_get_drvinfo, -}; - -/*====================================================================*/ - static int ei_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { diff --git a/drivers/net/sc92031.c b/drivers/net/sc92031.c index 8c4067af32b0..31b92f5f32cb 100644 --- a/drivers/net/sc92031.c +++ b/drivers/net/sc92031.c @@ -1251,16 +1251,6 @@ static int sc92031_ethtool_set_settings(struct net_device *dev, return 0; } -static void sc92031_ethtool_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *drvinfo) -{ - struct sc92031_priv *priv = netdev_priv(dev); - struct pci_dev *pdev = priv->pdev; - - strcpy(drvinfo->driver, SC92031_NAME); - strcpy(drvinfo->bus_info, pci_name(pdev)); -} - static void sc92031_ethtool_get_wol(struct net_device *dev, struct ethtool_wolinfo *wolinfo) { @@ -1382,7 +1372,6 @@ static void sc92031_ethtool_get_ethtool_stats(struct net_device *dev, static const struct ethtool_ops sc92031_ethtool_ops = { .get_settings = sc92031_ethtool_get_settings, .set_settings = sc92031_ethtool_set_settings, - .get_drvinfo = sc92031_ethtool_get_drvinfo, .get_wol = sc92031_ethtool_get_wol, .set_wol = sc92031_ethtool_set_wol, .nway_reset = sc92031_ethtool_nway_reset, diff --git a/drivers/net/tulip/xircom_cb.c b/drivers/net/tulip/xircom_cb.c index a439e93be22d..5a73752be2ca 100644 --- a/drivers/net/tulip/xircom_cb.c +++ b/drivers/net/tulip/xircom_cb.c @@ -29,7 +29,6 @@ #include #include #include -#include #include #include @@ -181,19 +180,6 @@ static void print_binary(unsigned int number) } #endif -static void netdev_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *info) -{ - struct xircom_private *private = netdev_priv(dev); - - strcpy(info->driver, "xircom_cb"); - strcpy(info->bus_info, pci_name(private->pdev)); -} - -static const struct ethtool_ops netdev_ethtool_ops = { - .get_drvinfo = netdev_get_drvinfo, -}; - static const struct net_device_ops netdev_ops = { .ndo_open = xircom_open, .ndo_stop = xircom_close, @@ -279,7 +265,6 @@ static int __devinit xircom_probe(struct pci_dev *pdev, const struct pci_device_ setup_descriptors(private); dev->netdev_ops = &netdev_ops; - SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops); pci_set_drvdata(pdev, dev); if (register_netdev(dev)) { diff --git a/drivers/net/usb/hso.c b/drivers/net/usb/hso.c index 6efca66b8766..4f123f869bdc 100644 --- a/drivers/net/usb/hso.c +++ b/drivers/net/usb/hso.c @@ -843,16 +843,7 @@ static netdev_tx_t hso_net_start_xmit(struct sk_buff *skb, return NETDEV_TX_OK; } -static void hso_get_drvinfo(struct net_device *net, struct ethtool_drvinfo *info) -{ - struct hso_net *odev = netdev_priv(net); - - strncpy(info->driver, driver_name, ETHTOOL_BUSINFO_LEN); - usb_make_path(odev->parent->usb, info->bus_info, sizeof info->bus_info); -} - static const struct ethtool_ops ops = { - .get_drvinfo = hso_get_drvinfo, .get_link = ethtool_op_get_link }; diff --git a/drivers/net/usb/kaweth.c b/drivers/net/usb/kaweth.c index 2b7b39cad1ce..5e98643a4a21 100644 --- a/drivers/net/usb/kaweth.c +++ b/drivers/net/usb/kaweth.c @@ -759,14 +759,6 @@ static int kaweth_close(struct net_device *net) return 0; } -static void kaweth_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) -{ - struct kaweth_device *kaweth = netdev_priv(dev); - - strlcpy(info->driver, driver_name, sizeof(info->driver)); - usb_make_path(kaweth->dev, info->bus_info, sizeof (info->bus_info)); -} - static u32 kaweth_get_link(struct net_device *dev) { struct kaweth_device *kaweth = netdev_priv(dev); @@ -775,7 +767,6 @@ static u32 kaweth_get_link(struct net_device *dev) } static const struct ethtool_ops ops = { - .get_drvinfo = kaweth_get_drvinfo, .get_link = kaweth_get_link }; diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 4598e9d2608f..bb6b67f6b0cc 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -705,19 +705,6 @@ static int virtnet_close(struct net_device *dev) return 0; } -static void virtnet_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *drvinfo) -{ - struct virtnet_info *vi = netdev_priv(dev); - struct virtio_device *vdev = vi->vdev; - - strncpy(drvinfo->driver, KBUILD_MODNAME, ARRAY_SIZE(drvinfo->driver)); - strncpy(drvinfo->version, "N/A", ARRAY_SIZE(drvinfo->version)); - strncpy(drvinfo->fw_version, "N/A", ARRAY_SIZE(drvinfo->fw_version)); - strncpy(drvinfo->bus_info, dev_name(&vdev->dev), - ARRAY_SIZE(drvinfo->bus_info)); -} - static int virtnet_set_tx_csum(struct net_device *dev, u32 data) { struct virtnet_info *vi = netdev_priv(dev); @@ -830,7 +817,6 @@ static void virtnet_vlan_rx_kill_vid(struct net_device *dev, u16 vid) } static const struct ethtool_ops virtnet_ethtool_ops = { - .get_drvinfo = virtnet_get_drvinfo, .set_tx_csum = virtnet_set_tx_csum, .set_sg = ethtool_op_set_sg, .set_tso = ethtool_op_set_tso, diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c index 88560d0ae50a..3bd9cf76517d 100644 --- a/drivers/net/wireless/ray_cs.c +++ b/drivers/net/wireless/ray_cs.c @@ -43,7 +43,6 @@ #include #include #include -#include #include #include @@ -80,8 +79,6 @@ static int ray_dev_config(struct net_device *dev, struct ifmap *map); static struct net_device_stats *ray_get_stats(struct net_device *dev); static int ray_dev_init(struct net_device *dev); -static const struct ethtool_ops netdev_ethtool_ops; - static int ray_open(struct net_device *dev); static netdev_tx_t ray_dev_start_xmit(struct sk_buff *skb, struct net_device *dev); @@ -333,7 +330,6 @@ static int ray_probe(struct pcmcia_device *p_dev) /* Raylink entries in the device structure */ dev->netdev_ops = &ray_netdev_ops; - SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops); dev->wireless_handlers = &ray_handler_def; #ifdef WIRELESS_SPY local->wireless_data.spy_data = &local->spy_data; @@ -1062,18 +1058,6 @@ AP to AP 1 1 dest AP src AP dest source } } /* end encapsulate_frame */ -/*===========================================================================*/ - -static void netdev_get_drvinfo(struct net_device *dev, - struct ethtool_drvinfo *info) -{ - strcpy(info->driver, "ray_cs"); -} - -static const struct ethtool_ops netdev_ethtool_ops = { - .get_drvinfo = netdev_get_drvinfo, -}; - /*====================================================================*/ /*------------------------------------------------------------------*/ diff --git a/drivers/net/wireless/wl3501_cs.c b/drivers/net/wireless/wl3501_cs.c index a1cc2d498a1c..420e9e986a18 100644 --- a/drivers/net/wireless/wl3501_cs.c +++ b/drivers/net/wireless/wl3501_cs.c @@ -29,7 +29,6 @@ #include #include -#include #include #include #include @@ -1411,15 +1410,6 @@ static struct iw_statistics *wl3501_get_wireless_stats(struct net_device *dev) return wstats; } -static void wl3501_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) -{ - strlcpy(info->driver, "wl3501_cs", sizeof(info->driver)); -} - -static const struct ethtool_ops ops = { - .get_drvinfo = wl3501_get_drvinfo -}; - /** * wl3501_detach - deletes a driver "instance" * @link - FILL_IN @@ -1905,7 +1895,6 @@ static int wl3501_probe(struct pcmcia_device *p_dev) this->p_dev = p_dev; dev->wireless_data = &this->wireless_data; dev->wireless_handlers = &wl3501_handler_def; - SET_ETHTOOL_OPS(dev, &ops); netif_stop_queue(dev); p_dev->priv = dev; diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 7a85367b3c2f..d2c4da5a6a4f 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -205,18 +205,24 @@ static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo info; const struct ethtool_ops *ops = dev->ethtool_ops; - if (!ops->get_drvinfo) - return -EOPNOTSUPP; - memset(&info, 0, sizeof(info)); info.cmd = ETHTOOL_GDRVINFO; - ops->get_drvinfo(dev, &info); + if (ops && ops->get_drvinfo) { + ops->get_drvinfo(dev, &info); + } else if (dev->dev.parent && dev->dev.parent->driver) { + strlcpy(info.bus_info, dev_name(dev->dev.parent), + sizeof(info.bus_info)); + strlcpy(info.driver, dev->dev.parent->driver->name, + sizeof(info.driver)); + } else { + return -EOPNOTSUPP; + } /* * this method of obtaining string set info is deprecated; * Use ETHTOOL_GSSET_INFO instead. */ - if (ops->get_sset_count) { + if (ops && ops->get_sset_count) { int rc; rc = ops->get_sset_count(dev, ETH_SS_TEST); @@ -229,9 +235,9 @@ static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev, if (rc >= 0) info.n_priv_flags = rc; } - if (ops->get_regs_len) + if (ops && ops->get_regs_len) info.regdump_len = ops->get_regs_len(dev); - if (ops->get_eeprom_len) + if (ops && ops->get_eeprom_len) info.eedump_len = ops->get_eeprom_len(dev); if (copy_to_user(useraddr, &info, sizeof(info))) @@ -1402,12 +1408,19 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) if (!dev || !netif_device_present(dev)) return -ENODEV; - if (!dev->ethtool_ops) - return -EOPNOTSUPP; - if (copy_from_user(ðcmd, useraddr, sizeof(ethcmd))) return -EFAULT; + if (!dev->ethtool_ops) { + /* ETHTOOL_GDRVINFO does not require any driver support. + * It is also unprivileged and does not change anything, + * so we can take a shortcut to it. */ + if (ethcmd == ETHTOOL_GDRVINFO) + return ethtool_get_drvinfo(dev, useraddr); + else + return -EOPNOTSUPP; + } + /* Allow some commands to be done by anyone */ switch (ethcmd) { case ETHTOOL_GDRVINFO: -- cgit v1.2.3 From 2244d07bfa2097cb00600da91c715a8aa547917e Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Tue, 17 Aug 2010 08:59:14 +0000 Subject: net: simplify flags for tx timestamping This patch removes the abstraction introduced by the union skb_shared_tx in the shared skb data. The access of the different union elements at several places led to some confusion about accessing the shared tx_flags e.g. in skb_orphan_try(). http://marc.info/?l=linux-netdev&m=128084897415886&w=2 Signed-off-by: Oliver Hartkopp Signed-off-by: David S. Miller --- Documentation/networking/timestamping.txt | 22 +++++++++------- drivers/net/bfin_mac.c | 10 +++---- drivers/net/gianfar.c | 15 +++++------ drivers/net/igb/igb.h | 2 +- drivers/net/igb/igb_main.c | 11 ++++---- include/linux/skbuff.h | 44 +++++++++++-------------------- include/net/ip.h | 2 +- include/net/sock.h | 8 ++---- net/can/raw.c | 4 +-- net/core/dev.c | 6 ++--- net/core/skbuff.c | 2 +- net/ipv4/icmp.c | 4 +-- net/ipv4/ip_output.c | 6 ++--- net/ipv4/raw.c | 2 +- net/ipv4/udp.c | 4 +-- net/packet/af_packet.c | 4 +-- net/socket.c | 9 +++---- 17 files changed, 68 insertions(+), 87 deletions(-) (limited to 'net/core') diff --git a/Documentation/networking/timestamping.txt b/Documentation/networking/timestamping.txt index e8c8f4f06c67..98097d8cb910 100644 --- a/Documentation/networking/timestamping.txt +++ b/Documentation/networking/timestamping.txt @@ -172,15 +172,19 @@ struct skb_shared_hwtstamps { }; Time stamps for outgoing packets are to be generated as follows: -- In hard_start_xmit(), check if skb_tx(skb)->hardware is set no-zero. - If yes, then the driver is expected to do hardware time stamping. +- In hard_start_xmit(), check if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) + is set no-zero. If yes, then the driver is expected to do hardware time + stamping. - If this is possible for the skb and requested, then declare - that the driver is doing the time stamping by setting the field - skb_tx(skb)->in_progress non-zero. You might want to keep a pointer - to the associated skb for the next step and not free the skb. A driver - not supporting hardware time stamping doesn't do that. A driver must - never touch sk_buff::tstamp! It is used to store software generated - time stamps by the network subsystem. + that the driver is doing the time stamping by setting the flag + SKBTX_IN_PROGRESS in skb_shinfo(skb)->tx_flags , e.g. with + + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; + + You might want to keep a pointer to the associated skb for the next step + and not free the skb. A driver not supporting hardware time stamping doesn't + do that. A driver must never touch sk_buff::tstamp! It is used to store + software generated time stamps by the network subsystem. - As soon as the driver has sent the packet and/or obtained a hardware time stamp for it, it passes the time stamp back by calling skb_hwtstamp_tx() with the original skb, the raw @@ -191,6 +195,6 @@ Time stamps for outgoing packets are to be generated as follows: this would occur at a later time in the processing pipeline than other software time stamping and therefore could lead to unexpected deltas between time stamps. -- If the driver did not call set skb_tx(skb)->in_progress, then +- If the driver did not set the SKBTX_IN_PROGRESS flag (see above), then dev_hard_start_xmit() checks whether software time stamping is wanted as fallback and potentially generates the time stamp. diff --git a/drivers/net/bfin_mac.c b/drivers/net/bfin_mac.c index 012613fde3f4..7a0e4156fade 100644 --- a/drivers/net/bfin_mac.c +++ b/drivers/net/bfin_mac.c @@ -803,15 +803,14 @@ static void bfin_dump_hwtamp(char *s, ktime_t *hw, ktime_t *ts, struct timecompa static void bfin_tx_hwtstamp(struct net_device *netdev, struct sk_buff *skb) { struct bfin_mac_local *lp = netdev_priv(netdev); - union skb_shared_tx *shtx = skb_tx(skb); - if (shtx->hardware) { + if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) { int timeout_cnt = MAX_TIMEOUT_CNT; /* When doing time stamping, keep the connection to the socket * a while longer */ - shtx->in_progress = 1; + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; /* * The timestamping is done at the EMAC module's MII/RMII interface @@ -991,7 +990,6 @@ static int bfin_mac_hard_start_xmit(struct sk_buff *skb, struct bfin_mac_local *lp = netdev_priv(dev); u16 *data; u32 data_align = (unsigned long)(skb->data) & 0x3; - union skb_shared_tx *shtx = skb_tx(skb); current_tx_ptr->skb = skb; @@ -1005,7 +1003,7 @@ static int bfin_mac_hard_start_xmit(struct sk_buff *skb, * of this field are the length of the packet payload in bytes and the higher * 4 bits are the timestamping enable field. */ - if (shtx->hardware) + if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) *data |= 0x1000; current_tx_ptr->desc_a.start_addr = (u32)data; @@ -1015,7 +1013,7 @@ static int bfin_mac_hard_start_xmit(struct sk_buff *skb, } else { *((u16 *)(current_tx_ptr->packet)) = (u16)(skb->len); /* enable timestamping for the sent packet */ - if (shtx->hardware) + if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) *((u16 *)(current_tx_ptr->packet)) |= 0x1000; memcpy((u8 *)(current_tx_ptr->packet + 2), skb->data, skb->len); diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c index 3d9f958ebd2c..e6048d6ab0ea 100644 --- a/drivers/net/gianfar.c +++ b/drivers/net/gianfar.c @@ -2048,7 +2048,6 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev) u32 bufaddr; unsigned long flags; unsigned int nr_frags, nr_txbds, length; - union skb_shared_tx *shtx; /* * TOE=1 frames larger than 2500 bytes may see excess delays @@ -2069,10 +2068,10 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev) txq = netdev_get_tx_queue(dev, rq); base = tx_queue->tx_bd_base; regs = tx_queue->grp->regs; - shtx = skb_tx(skb); /* check if time stamp should be generated */ - if (unlikely(shtx->hardware && priv->hwts_tx_en)) + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP && + priv->hwts_tx_en)) do_tstamp = 1; /* make space for additional header when fcb is needed */ @@ -2174,7 +2173,7 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev) /* Setup tx hardware time stamping if requested */ if (unlikely(do_tstamp)) { - shtx->in_progress = 1; + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; if (fcb == NULL) fcb = gfar_add_fcb(skb); fcb->ptp = 1; @@ -2446,7 +2445,6 @@ static int gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue) int howmany = 0; u32 lstatus; size_t buflen; - union skb_shared_tx *shtx; rx_queue = priv->rx_queue[tx_queue->qindex]; bdp = tx_queue->dirty_tx; @@ -2461,8 +2459,7 @@ static int gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue) * When time stamping, one additional TxBD must be freed. * Also, we need to dma_unmap_single() the TxPAL. */ - shtx = skb_tx(skb); - if (unlikely(shtx->in_progress)) + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) nr_txbds = frags + 2; else nr_txbds = frags + 1; @@ -2476,7 +2473,7 @@ static int gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue) (lstatus & BD_LENGTH_MASK)) break; - if (unlikely(shtx->in_progress)) { + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) { next = next_txbd(bdp, base, tx_ring_size); buflen = next->length + GMAC_FCB_LEN; } else @@ -2485,7 +2482,7 @@ static int gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue) dma_unmap_single(&priv->ofdev->dev, bdp->bufPtr, buflen, DMA_TO_DEVICE); - if (unlikely(shtx->in_progress)) { + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) { struct skb_shared_hwtstamps shhwtstamps; u64 *ns = (u64*) (((u32)skb->data + 0x10) & ~0x7); memset(&shhwtstamps, 0, sizeof(shhwtstamps)); diff --git a/drivers/net/igb/igb.h b/drivers/net/igb/igb.h index 6e63d9a7fc75..44e0ff1494e0 100644 --- a/drivers/net/igb/igb.h +++ b/drivers/net/igb/igb.h @@ -143,7 +143,7 @@ struct igb_buffer { u16 next_to_watch; unsigned int bytecount; u16 gso_segs; - union skb_shared_tx shtx; + u8 tx_flags; u8 mapped_as_page; }; /* RX */ diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c index 9b4e5895f5f9..985e37cf17b6 100644 --- a/drivers/net/igb/igb_main.c +++ b/drivers/net/igb/igb_main.c @@ -3954,7 +3954,7 @@ static inline int igb_tx_map_adv(struct igb_ring *tx_ring, struct sk_buff *skb, } tx_ring->buffer_info[i].skb = skb; - tx_ring->buffer_info[i].shtx = skb_shinfo(skb)->tx_flags; + tx_ring->buffer_info[i].tx_flags = skb_shinfo(skb)->tx_flags; /* multiply data chunks by size of headers */ tx_ring->buffer_info[i].bytecount = ((gso_segs - 1) * hlen) + skb->len; tx_ring->buffer_info[i].gso_segs = gso_segs; @@ -4088,7 +4088,6 @@ netdev_tx_t igb_xmit_frame_ring_adv(struct sk_buff *skb, u32 tx_flags = 0; u16 first; u8 hdr_len = 0; - union skb_shared_tx *shtx = skb_tx(skb); /* need: 1 descriptor per page, * + 2 desc gap to keep tail from touching head, @@ -4100,8 +4099,8 @@ netdev_tx_t igb_xmit_frame_ring_adv(struct sk_buff *skb, return NETDEV_TX_BUSY; } - if (unlikely(shtx->hardware)) { - shtx->in_progress = 1; + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) { + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; tx_flags |= IGB_TX_FLAGS_TSTAMP; } @@ -5319,7 +5318,7 @@ static void igb_tx_hwtstamp(struct igb_q_vector *q_vector, struct igb_buffer *bu u64 regval; /* if skb does not support hw timestamp or TX stamp not valid exit */ - if (likely(!buffer_info->shtx.hardware) || + if (likely(!(buffer_info->tx_flags & SKBTX_HW_TSTAMP)) || !(rd32(E1000_TSYNCTXCTL) & E1000_TSYNCTXCTL_VALID)) return; @@ -5500,7 +5499,7 @@ static void igb_rx_hwtstamp(struct igb_q_vector *q_vector, u32 staterr, * values must belong to this one here and therefore we don't need to * compare any of the additional attributes stored for it. * - * If nothing went wrong, then it should have a skb_shared_tx that we + * If nothing went wrong, then it should have a shared tx_flags that we * can turn into a skb_shared_hwtstamps. */ if (staterr & E1000_RXDADV_STAT_TSIP) { diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index d8050382b189..f067c95cf18a 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -163,26 +163,19 @@ struct skb_shared_hwtstamps { ktime_t syststamp; }; -/** - * struct skb_shared_tx - instructions for time stamping of outgoing packets - * @hardware: generate hardware time stamp - * @software: generate software time stamp - * @in_progress: device driver is going to provide - * hardware time stamp - * @prevent_sk_orphan: make sk reference available on driver level - * @flags: all shared_tx flags - * - * These flags are attached to packets as part of the - * &skb_shared_info. Use skb_tx() to get a pointer. - */ -union skb_shared_tx { - struct { - __u8 hardware:1, - software:1, - in_progress:1, - prevent_sk_orphan:1; - }; - __u8 flags; +/* Definitions for tx_flags in struct skb_shared_info */ +enum { + /* generate hardware time stamp */ + SKBTX_HW_TSTAMP = 1 << 0, + + /* generate software time stamp */ + SKBTX_SW_TSTAMP = 1 << 1, + + /* device driver is going to provide hardware time stamp */ + SKBTX_IN_PROGRESS = 1 << 2, + + /* ensure the originating sk reference is available on driver level */ + SKBTX_DRV_NEEDS_SK_REF = 1 << 3, }; /* This data is invariant across clones and lives at @@ -195,7 +188,7 @@ struct skb_shared_info { unsigned short gso_segs; unsigned short gso_type; __be32 ip6_frag_id; - union skb_shared_tx tx_flags; + __u8 tx_flags; struct sk_buff *frag_list; struct skb_shared_hwtstamps hwtstamps; @@ -587,11 +580,6 @@ static inline struct skb_shared_hwtstamps *skb_hwtstamps(struct sk_buff *skb) return &skb_shinfo(skb)->hwtstamps; } -static inline union skb_shared_tx *skb_tx(struct sk_buff *skb) -{ - return &skb_shinfo(skb)->tx_flags; -} - /** * skb_queue_empty - check if a queue is empty * @list: queue head @@ -1996,8 +1984,8 @@ extern void skb_tstamp_tx(struct sk_buff *orig_skb, static inline void sw_tx_timestamp(struct sk_buff *skb) { - union skb_shared_tx *shtx = skb_tx(skb); - if (shtx->software && !shtx->in_progress) + if (skb_shinfo(skb)->tx_flags & SKBTX_SW_TSTAMP && + !(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) skb_tstamp_tx(skb, NULL); } diff --git a/include/net/ip.h b/include/net/ip.h index 890f9725d681..7691aca133db 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -53,7 +53,7 @@ struct ipcm_cookie { __be32 addr; int oif; struct ip_options *opt; - union skb_shared_tx shtx; + __u8 tx_flags; }; #define IPCB(skb) ((struct inet_skb_parm*)((skb)->cb)) diff --git a/include/net/sock.h b/include/net/sock.h index ac53bfbdfe16..100e43bf95fb 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1669,17 +1669,13 @@ static inline void sock_recv_ts_and_drops(struct msghdr *msg, struct sock *sk, /** * sock_tx_timestamp - checks whether the outgoing packet is to be time stamped - * @msg: outgoing packet * @sk: socket sending this packet - * @shtx: filled with instructions for time stamping + * @tx_flags: filled with instructions for time stamping * * Currently only depends on SOCK_TIMESTAMPING* flags. Returns error code if * parameters are invalid. */ -extern int sock_tx_timestamp(struct msghdr *msg, - struct sock *sk, - union skb_shared_tx *shtx); - +extern int sock_tx_timestamp(struct sock *sk, __u8 *tx_flags); /** * sk_eat_skb - Release a skb if it is no longer needed diff --git a/net/can/raw.c b/net/can/raw.c index a10e3338f084..7d77e67e57af 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -647,12 +647,12 @@ static int raw_sendmsg(struct kiocb *iocb, struct socket *sock, err = memcpy_fromiovec(skb_put(skb, size), msg->msg_iov, size); if (err < 0) goto free_skb; - err = sock_tx_timestamp(msg, sk, skb_tx(skb)); + err = sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); if (err < 0) goto free_skb; /* to be able to check the received tx sock reference in raw_rcv() */ - skb_tx(skb)->prevent_sk_orphan = 1; + skb_shinfo(skb)->tx_flags |= SKBTX_DRV_NEEDS_SK_REF; skb->dev = dev; skb->sk = sk; diff --git a/net/core/dev.c b/net/core/dev.c index 586a11cb4398..c1dc8a95f6ff 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1902,14 +1902,14 @@ static int dev_gso_segment(struct sk_buff *skb) /* * Try to orphan skb early, right before transmission by the device. - * We cannot orphan skb if tx timestamp is requested, since - * drivers need to call skb_tstamp_tx() to send the timestamp. + * We cannot orphan skb if tx timestamp is requested or the sk-reference + * is needed on driver level for other reasons, e.g. see net/can/raw.c */ static inline void skb_orphan_try(struct sk_buff *skb) { struct sock *sk = skb->sk; - if (sk && !skb_tx(skb)->flags) { + if (sk && !skb_shinfo(skb)->tx_flags) { /* skb_tx_hash() wont be able to get sk. * We copy sk_hash into skb->rxhash */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 3a2513f0d0c3..99ef721f773d 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -3016,7 +3016,7 @@ void skb_tstamp_tx(struct sk_buff *orig_skb, } else { /* * no hardware time stamps available, - * so keep the skb_shared_tx and only + * so keep the shared tx_flags and only * store software time stamp */ skb->tstamp = ktime_get_real(); diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index a0d847c7cba5..96bc7f9475a3 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -379,7 +379,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) inet->tos = ip_hdr(skb)->tos; daddr = ipc.addr = rt->rt_src; ipc.opt = NULL; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; if (icmp_param->replyopts.optlen) { ipc.opt = &icmp_param->replyopts; if (ipc.opt->srr) @@ -538,7 +538,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) inet_sk(sk)->tos = tos; ipc.addr = iph->saddr; ipc.opt = &icmp_param.replyopts; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; { struct flowi fl = { diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 04b69896df5f..e807492f1777 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -953,7 +953,7 @@ alloc_new_skb: else /* only the initial fragment is time stamped */ - ipc->shtx.flags = 0; + ipc->tx_flags = 0; } if (skb == NULL) goto error; @@ -964,7 +964,7 @@ alloc_new_skb: skb->ip_summed = csummode; skb->csum = 0; skb_reserve(skb, hh_len); - *skb_tx(skb) = ipc->shtx; + skb_shinfo(skb)->tx_flags = ipc->tx_flags; /* * Find where to start putting bytes. @@ -1384,7 +1384,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar daddr = ipc.addr = rt->rt_src; ipc.opt = NULL; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; if (replyopts.opt.optlen) { ipc.opt = &replyopts.opt; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 009a7b2aa1ef..1f85ef289895 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -505,7 +505,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.addr = inet->inet_saddr; ipc.opt = NULL; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; ipc.oif = sk->sk_bound_dev_if; if (msg->msg_controllen) { diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 32e0bef60d0a..86e757e162ee 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -797,7 +797,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, return -EOPNOTSUPP; ipc.opt = NULL; - ipc.shtx.flags = 0; + ipc.tx_flags = 0; if (up->pending) { /* @@ -845,7 +845,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.addr = inet->inet_saddr; ipc.oif = sk->sk_bound_dev_if; - err = sock_tx_timestamp(msg, sk, &ipc.shtx); + err = sock_tx_timestamp(sk, &ipc.tx_flags); if (err) return err; if (msg->msg_controllen) { diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 9a17f28b1253..3616f27b9d46 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -488,7 +488,7 @@ retry: skb->dev = dev; skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; - err = sock_tx_timestamp(msg, sk, skb_tx(skb)); + err = sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); if (err < 0) goto out_unlock; @@ -1209,7 +1209,7 @@ static int packet_snd(struct socket *sock, err = skb_copy_datagram_from_iovec(skb, offset, msg->msg_iov, 0, len); if (err) goto out_free; - err = sock_tx_timestamp(msg, sk, skb_tx(skb)); + err = sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); if (err < 0) goto out_free; diff --git a/net/socket.c b/net/socket.c index 2270b941bcc7..7848d12f5e4d 100644 --- a/net/socket.c +++ b/net/socket.c @@ -535,14 +535,13 @@ void sock_release(struct socket *sock) } EXPORT_SYMBOL(sock_release); -int sock_tx_timestamp(struct msghdr *msg, struct sock *sk, - union skb_shared_tx *shtx) +int sock_tx_timestamp(struct sock *sk, __u8 *tx_flags) { - shtx->flags = 0; + *tx_flags = 0; if (sock_flag(sk, SOCK_TIMESTAMPING_TX_HARDWARE)) - shtx->hardware = 1; + *tx_flags |= SKBTX_HW_TSTAMP; if (sock_flag(sk, SOCK_TIMESTAMPING_TX_SOFTWARE)) - shtx->software = 1; + *tx_flags |= SKBTX_SW_TSTAMP; return 0; } EXPORT_SYMBOL(sock_tx_timestamp); -- cgit v1.2.3 From 2d47b45951af087c1a4439c559309b0bf90a0718 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 17 Aug 2010 19:00:56 +0000 Subject: net: rps: reset network header before calling skb_get_rxhash() skb_get_rxhash() assumes the network header pointer of the skb is set properly after the commit: commit bfb564e7391340638afe4ad67744a8f3858e7566 Author: Krishna Kumar Date: Wed Aug 4 06:15:52 2010 +0000 core: Factor out flow calculation from get_rps_cpu Signed-off-by: Changli Gao Signed-off-by: David S. Miller --- net/core/dev.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index c1dc8a95f6ff..cf87fde3a29b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2372,6 +2372,7 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, if (!rxqueue->rps_map && !rxqueue->rps_flow_table) goto done; + skb_reset_network_header(skb); if (!skb_get_rxhash(skb)) goto done; -- cgit v1.2.3 From dbe5775bbc00116ed5699babfe17c54f32eb34c3 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 17 Aug 2010 19:01:38 +0000 Subject: net: rps: skip fragment when computing rxhash Fragmented IP packets may have no transfer header, so when computing rxhash, we should skip them. Signed-off-by: Changli Gao Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index cf87fde3a29b..7e97e891636e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2284,7 +2284,10 @@ __u32 __skb_get_rxhash(struct sk_buff *skb) goto done; ip = (struct iphdr *) skb->data + nhoff; - ip_proto = ip->protocol; + if (ip->frag_off & htons(IP_MF | IP_OFFSET)) + ip_proto = 0; + else + ip_proto = ip->protocol; addr1 = (__force u32) ip->saddr; addr2 = (__force u32) ip->daddr; ihl = ip->ihl; -- cgit v1.2.3 From 12fcdefb3643607c47f39906a49056cf608bb545 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 17 Aug 2010 19:04:32 +0000 Subject: net: rps: use proto_ports_offset() to handle the AH message correctly The SPI isn't at the beginning of an AH message. Signed-off-by: Changli Gao Signed-off-by: David S. Miller --- net/core/dev.c | 23 +++++++---------------- 1 file changed, 7 insertions(+), 16 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 7e97e891636e..da584f5ab3d2 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2266,7 +2266,7 @@ static inline void ____napi_schedule(struct softnet_data *sd, */ __u32 __skb_get_rxhash(struct sk_buff *skb) { - int nhoff, hash = 0; + int nhoff, hash = 0, poff; struct ipv6hdr *ip6; struct iphdr *ip; u8 ip_proto; @@ -2306,24 +2306,15 @@ __u32 __skb_get_rxhash(struct sk_buff *skb) goto done; } - switch (ip_proto) { - case IPPROTO_TCP: - case IPPROTO_UDP: - case IPPROTO_DCCP: - case IPPROTO_ESP: - case IPPROTO_AH: - case IPPROTO_SCTP: - case IPPROTO_UDPLITE: - if (pskb_may_pull(skb, (ihl * 4) + 4 + nhoff)) { - ports.v32 = * (__force u32 *) (skb->data + nhoff + - (ihl * 4)); + ports.v32 = 0; + poff = proto_ports_offset(ip_proto); + if (poff >= 0) { + nhoff += ihl * 4 + poff; + if (pskb_may_pull(skb, nhoff + 4)) { + ports.v32 = * (__force u32 *) (skb->data + nhoff); if (ports.v16[1] < ports.v16[0]) swap(ports.v16[0], ports.v16[1]); - break; } - default: - ports.v32 = 0; - break; } /* get a consistent hash (same value on both flow directions) */ -- cgit v1.2.3 From 1003489e06c04d807c783a8958f2ccc9aed7a244 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Sat, 21 Aug 2010 06:13:28 +0000 Subject: net: rps: fix the wrong network header pointer __skb_get_rxhash() was broken after the commit: commit bfb564e7391340638afe4ad67744a8f3858e7566 Author: Krishna Kumar Date: Wed Aug 4 06:15:52 2010 +0000 core: Factor out flow calculation from get_rps_cpu Signed-off-by: Changli Gao Signed-off-by: David S. Miller --- net/core/dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index da584f5ab3d2..4d74d26b8e1a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2283,7 +2283,7 @@ __u32 __skb_get_rxhash(struct sk_buff *skb) if (!pskb_may_pull(skb, sizeof(*ip) + nhoff)) goto done; - ip = (struct iphdr *) skb->data + nhoff; + ip = (struct iphdr *) (skb->data + nhoff); if (ip->frag_off & htons(IP_MF | IP_OFFSET)) ip_proto = 0; else @@ -2296,7 +2296,7 @@ __u32 __skb_get_rxhash(struct sk_buff *skb) if (!pskb_may_pull(skb, sizeof(*ip6) + nhoff)) goto done; - ip6 = (struct ipv6hdr *) skb->data + nhoff; + ip6 = (struct ipv6hdr *) (skb->data + nhoff); ip_proto = ip6->nexthdr; addr1 = (__force u32) ip6->saddr.s6_addr32[3]; addr2 = (__force u32) ip6->daddr.s6_addr32[3]; -- cgit v1.2.3 From 05532121da0728eaedac2a0a5c3cecad3a95d765 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Sun, 22 Aug 2010 21:03:33 -0700 Subject: net: 802.1q: make vlan_hwaccel_do_receive() return void vlan_hwaccel_do_receive() always returns 0, so make it return void. Signed-off-by: Changli Gao Signed-off-by: David S. Miller --- include/linux/if_vlan.h | 5 ++--- net/8021q/vlan_core.c | 3 +-- net/core/dev.c | 4 ++-- 3 files changed, 5 insertions(+), 7 deletions(-) (limited to 'net/core') diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 3d870fda8c4f..a52320751bfc 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -119,7 +119,7 @@ extern u16 vlan_dev_vlan_id(const struct net_device *dev); extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, u16 vlan_tci, int polling); -extern int vlan_hwaccel_do_receive(struct sk_buff *skb); +extern void vlan_hwaccel_do_receive(struct sk_buff *skb); extern gro_result_t vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp, unsigned int vlan_tci, struct sk_buff *skb); @@ -147,9 +147,8 @@ static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, return NET_XMIT_SUCCESS; } -static inline int vlan_hwaccel_do_receive(struct sk_buff *skb) +static inline void vlan_hwaccel_do_receive(struct sk_buff *skb) { - return 0; } static inline gro_result_t diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index 01ddb0472f86..07eeb5b99dce 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -35,7 +35,7 @@ drop: } EXPORT_SYMBOL(__vlan_hwaccel_rx); -int vlan_hwaccel_do_receive(struct sk_buff *skb) +void vlan_hwaccel_do_receive(struct sk_buff *skb) { struct net_device *dev = skb->dev; struct vlan_rx_stats *rx_stats; @@ -69,7 +69,6 @@ int vlan_hwaccel_do_receive(struct sk_buff *skb) break; } u64_stats_update_end(&rx_stats->syncp); - return 0; } struct net_device *vlan_dev_real_dev(const struct net_device *dev) diff --git a/net/core/dev.c b/net/core/dev.c index 7cd5237d9822..d569f88bcf80 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2841,8 +2841,8 @@ static int __netif_receive_skb(struct sk_buff *skb) if (!netdev_tstamp_prequeue) net_timestamp_check(skb); - if (vlan_tx_tag_present(skb) && vlan_hwaccel_do_receive(skb)) - return NET_RX_SUCCESS; + if (vlan_tx_tag_present(skb)) + vlan_hwaccel_do_receive(skb); /* if we've gotten here through NAPI, check netpoll */ if (netpoll_receive_skb(skb)) -- cgit v1.2.3 From 21dc330157454046dd7c494961277d76e1c957fe Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 23 Aug 2010 00:13:46 -0700 Subject: net: Rename skb_has_frags to skb_has_frag_list SKBs can be "fragmented" in two ways, via a page array (called skb_shinfo(skb)->frags[]) and via a list of SKBs (called skb_shinfo(skb)->frag_list). Since skb_has_frags() tests the latter, it's name is confusing since it sounds more like it's testing the former. Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- include/linux/skbuff.h | 4 ++-- net/core/dev.c | 4 ++-- net/core/skbuff.c | 18 +++++++++--------- net/ipv4/ip_fragment.c | 2 +- net/ipv4/ip_output.c | 2 +- net/ipv6/ip6_output.c | 2 +- net/ipv6/netfilter/nf_conntrack_reasm.c | 2 +- net/ipv6/reassembly.c | 2 +- 9 files changed, 19 insertions(+), 19 deletions(-) (limited to 'net/core') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 46c36ffe20ee..ce2de8b64083 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2191,7 +2191,7 @@ static inline int net_gso_ok(int features, int gso_type) static inline int skb_gso_ok(struct sk_buff *skb, int features) { return net_gso_ok(features, skb_shinfo(skb)->gso_type) && - (!skb_has_frags(skb) || (features & NETIF_F_FRAGLIST)); + (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST)); } static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index f067c95cf18a..f900ffcd847e 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1120,7 +1120,7 @@ extern void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page, int off, int size); #define SKB_PAGE_ASSERT(skb) BUG_ON(skb_shinfo(skb)->nr_frags) -#define SKB_FRAG_ASSERT(skb) BUG_ON(skb_has_frags(skb)) +#define SKB_FRAG_ASSERT(skb) BUG_ON(skb_has_frag_list(skb)) #define SKB_LINEAR_ASSERT(skb) BUG_ON(skb_is_nonlinear(skb)) #ifdef NET_SKBUFF_DATA_USES_OFFSET @@ -1784,7 +1784,7 @@ static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len) skb = skb->prev) -static inline bool skb_has_frags(const struct sk_buff *skb) +static inline bool skb_has_frag_list(const struct sk_buff *skb) { return skb_shinfo(skb)->frag_list != NULL; } diff --git a/net/core/dev.c b/net/core/dev.c index d569f88bcf80..859e30ff044a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1930,7 +1930,7 @@ static inline int skb_needs_linearize(struct sk_buff *skb, struct net_device *dev) { return skb_is_nonlinear(skb) && - ((skb_has_frags(skb) && !(dev->features & NETIF_F_FRAGLIST)) || + ((skb_has_frag_list(skb) && !(dev->features & NETIF_F_FRAGLIST)) || (skb_shinfo(skb)->nr_frags && (!(dev->features & NETIF_F_SG) || illegal_highdma(dev, skb)))); } @@ -3090,7 +3090,7 @@ enum gro_result dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) if (!(skb->dev->features & NETIF_F_GRO) || netpoll_rx_on(skb)) goto normal; - if (skb_is_gso(skb) || skb_has_frags(skb)) + if (skb_is_gso(skb) || skb_has_frag_list(skb)) goto normal; rcu_read_lock(); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 99ef721f773d..e2535fb4985d 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -340,7 +340,7 @@ static void skb_release_data(struct sk_buff *skb) put_page(skb_shinfo(skb)->frags[i].page); } - if (skb_has_frags(skb)) + if (skb_has_frag_list(skb)) skb_drop_fraglist(skb); kfree(skb->head); @@ -759,7 +759,7 @@ struct sk_buff *pskb_copy(struct sk_buff *skb, gfp_t gfp_mask) skb_shinfo(n)->nr_frags = i; } - if (skb_has_frags(skb)) { + if (skb_has_frag_list(skb)) { skb_shinfo(n)->frag_list = skb_shinfo(skb)->frag_list; skb_clone_fraglist(n); } @@ -822,7 +822,7 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) get_page(skb_shinfo(skb)->frags[i].page); - if (skb_has_frags(skb)) + if (skb_has_frag_list(skb)) skb_clone_fraglist(skb); skb_release_data(skb); @@ -1099,7 +1099,7 @@ drop_pages: for (; i < nfrags; i++) put_page(skb_shinfo(skb)->frags[i].page); - if (skb_has_frags(skb)) + if (skb_has_frag_list(skb)) skb_drop_fraglist(skb); goto done; } @@ -1194,7 +1194,7 @@ unsigned char *__pskb_pull_tail(struct sk_buff *skb, int delta) /* Optimization: no fragments, no reasons to preestimate * size of pulled pages. Superb. */ - if (!skb_has_frags(skb)) + if (!skb_has_frag_list(skb)) goto pull_pages; /* Estimate size of pulled pages. */ @@ -2323,7 +2323,7 @@ next_skb: st->frag_data = NULL; } - if (st->root_skb == st->cur_skb && skb_has_frags(st->root_skb)) { + if (st->root_skb == st->cur_skb && skb_has_frag_list(st->root_skb)) { st->cur_skb = skb_shinfo(st->root_skb)->frag_list; st->frag_idx = 0; goto next_skb; @@ -2889,7 +2889,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer) return -ENOMEM; /* Easy case. Most of packets will go this way. */ - if (!skb_has_frags(skb)) { + if (!skb_has_frag_list(skb)) { /* A little of trouble, not enough of space for trailer. * This should not happen, when stack is tuned to generate * good frames. OK, on miss we reallocate and reserve even more @@ -2924,7 +2924,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer) if (skb1->next == NULL && tailbits) { if (skb_shinfo(skb1)->nr_frags || - skb_has_frags(skb1) || + skb_has_frag_list(skb1) || skb_tailroom(skb1) < tailbits) ntail = tailbits + 128; } @@ -2933,7 +2933,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer) skb_cloned(skb1) || ntail || skb_shinfo(skb1)->nr_frags || - skb_has_frags(skb1)) { + skb_has_frag_list(skb1)) { struct sk_buff *skb2; /* Fuck, we are miserable poor guys... */ diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index b7c41654dde5..f4dc879e258e 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -542,7 +542,7 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev, /* If the first fragment is fragmented itself, we split * it to two chunks: the first with data and paged part * and the second, holding only fragments. */ - if (skb_has_frags(head)) { + if (skb_has_frag_list(head)) { struct sk_buff *clone; int i, plen = 0; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index e807492f1777..6d2753c7ffdd 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -487,7 +487,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) * LATER: this step can be merged to real generation of fragments, * we can switch to copy when see the first bad fragment. */ - if (skb_has_frags(skb)) { + if (skb_has_frag_list(skb)) { struct sk_buff *frag; int first_len = skb_pagelen(skb); int truesizes = 0; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index d40b330c0ee6..1838927a2243 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -637,7 +637,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) } mtu -= hlen + sizeof(struct frag_hdr); - if (skb_has_frags(skb)) { + if (skb_has_frag_list(skb)) { int first_len = skb_pagelen(skb); int truesizes = 0; diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 13ef5bc05cf5..089c598773c7 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -413,7 +413,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev) /* If the first fragment is fragmented itself, we split * it to two chunks: the first with data and paged part * and the second, holding only fragments. */ - if (skb_has_frags(head)) { + if (skb_has_frag_list(head)) { struct sk_buff *clone; int i, plen = 0; diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 545c4141b755..8aea3f3f18d7 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -499,7 +499,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev, /* If the first fragment is fragmented itself, we split * it to two chunks: the first with data and paged part * and the second, holding only fragments. */ - if (skb_has_frags(head)) { + if (skb_has_frag_list(head)) { struct sk_buff *clone; int i, plen = 0; -- cgit v1.2.3 From afdcba371f9748ac91608bb6c57f170aab7085b4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 23 Aug 2010 07:14:36 +0000 Subject: net: copy_rtnl_link_stats64() simplification No need to use a temporary struct rtnl_link_stats64 variable, just copy the source to skb buffer. Signed-off-by: Eric Dumazet Reviewed-by: Ben Hutchings Signed-off-by: David S. Miller --- net/core/rtnetlink.c | 31 +------------------------------ 1 file changed, 1 insertion(+), 30 deletions(-) (limited to 'net/core') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index f78d821bd935..b2a718dfd720 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -612,36 +612,7 @@ static void copy_rtnl_link_stats(struct rtnl_link_stats *a, static void copy_rtnl_link_stats64(void *v, const struct rtnl_link_stats64 *b) { - struct rtnl_link_stats64 a; - - a.rx_packets = b->rx_packets; - a.tx_packets = b->tx_packets; - a.rx_bytes = b->rx_bytes; - a.tx_bytes = b->tx_bytes; - a.rx_errors = b->rx_errors; - a.tx_errors = b->tx_errors; - a.rx_dropped = b->rx_dropped; - a.tx_dropped = b->tx_dropped; - - a.multicast = b->multicast; - a.collisions = b->collisions; - - a.rx_length_errors = b->rx_length_errors; - a.rx_over_errors = b->rx_over_errors; - a.rx_crc_errors = b->rx_crc_errors; - a.rx_frame_errors = b->rx_frame_errors; - a.rx_fifo_errors = b->rx_fifo_errors; - a.rx_missed_errors = b->rx_missed_errors; - - a.tx_aborted_errors = b->tx_aborted_errors; - a.tx_carrier_errors = b->tx_carrier_errors; - a.tx_fifo_errors = b->tx_fifo_errors; - a.tx_heartbeat_errors = b->tx_heartbeat_errors; - a.tx_window_errors = b->tx_window_errors; - - a.rx_compressed = b->rx_compressed; - a.tx_compressed = b->tx_compressed; - memcpy(v, &a, sizeof(a)); + memcpy(v, b, sizeof(*b)); } /* All VF info */ -- cgit v1.2.3 From 0fdc100bdc4b7ab61ed632962c76dfe539047296 Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Mon, 23 Aug 2010 10:24:18 +0000 Subject: ethtool: allow non-netadmin to query settings The SNMP daemon uses ethtool to determine the speed of network interfaces. This fails on Debian (and probably elsewhere) because for security SNMP daemon runs as non-root user (snmp). Note: A similar patch was rejected previously because of a concern about the possibility that on some hardware querying the ethtool settings requires access to the PHY and could slow the machine down. But the security risk of requiring SNMP daemon (and related services) to run as root far out weighs the risk of denial-of-service. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/core/ethtool.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/core') diff --git a/net/core/ethtool.c b/net/core/ethtool.c index d2c4da5a6a4f..970eb9817bbc 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -1423,6 +1423,7 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) /* Allow some commands to be done by anyone */ switch (ethcmd) { + case ETHTOOL_GSET: case ETHTOOL_GDRVINFO: case ETHTOOL_GMSGLVL: case ETHTOOL_GCOALESCE: -- cgit v1.2.3 From 40d0802b3eb47d57e2d57a5244a18cbbe9632e13 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Aug 2010 22:03:08 -0700 Subject: gro: __napi_gro_receive() optimizations compare_ether_header() can have a special implementation on 64 bit arches if CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS is defined. __napi_gro_receive() and vlan_gro_common() can avoid a conditional branch to perform device match. On x86_64, __napi_gro_receive() has now 38 instructions instead of 53 As gcc-4.4.3 still choose to not inline it, add inline keyword to this performance critical function. Signed-off-by: Eric Dumazet CC: Herbert Xu Signed-off-by: David S. Miller --- include/linux/etherdevice.h | 18 +++++++++++++++++- net/8021q/vlan_core.c | 9 ++++++--- net/core/dev.c | 10 ++++++---- 3 files changed, 29 insertions(+), 8 deletions(-) (limited to 'net/core') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index 2308fbb4523a..fb6aa6070921 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -237,13 +237,29 @@ static inline bool is_etherdev_addr(const struct net_device *dev, * entry points. */ -static inline int compare_ether_header(const void *a, const void *b) +static inline unsigned long compare_ether_header(const void *a, const void *b) { +#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64 + unsigned long fold; + + /* + * We want to compare 14 bytes: + * [a0 ... a13] ^ [b0 ... b13] + * Use two long XOR, ORed together, with an overlap of two bytes. + * [a0 a1 a2 a3 a4 a5 a6 a7 ] ^ [b0 b1 b2 b3 b4 b5 b6 b7 ] | + * [a6 a7 a8 a9 a10 a11 a12 a13] ^ [b6 b7 b8 b9 b10 b11 b12 b13] + * This means the [a6 a7] ^ [b6 b7] part is done two times. + */ + fold = *(unsigned long *)a ^ *(unsigned long *)b; + fold |= *(unsigned long *)(a + 6) ^ *(unsigned long *)(b + 6); + return fold; +#else u32 *a32 = (u32 *)((u8 *)a + 2); u32 *b32 = (u32 *)((u8 *)b + 2); return (*(u16 *)a ^ *(u16 *)b) | (a32[0] ^ b32[0]) | (a32[1] ^ b32[1]) | (a32[2] ^ b32[2]); +#endif } #endif /* _LINUX_ETHERDEVICE_H */ diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index 07eeb5b99dce..3438c01bbacf 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -105,9 +105,12 @@ vlan_gro_common(struct napi_struct *napi, struct vlan_group *grp, goto drop; for (p = napi->gro_list; p; p = p->next) { - NAPI_GRO_CB(p)->same_flow = - p->dev == skb->dev && !compare_ether_header( - skb_mac_header(p), skb_gro_mac_header(skb)); + unsigned long diffs; + + diffs = (unsigned long)p->dev ^ (unsigned long)skb->dev; + diffs |= compare_ether_header(skb_mac_header(p), + skb_gro_mac_header(skb)); + NAPI_GRO_CB(p)->same_flow = !diffs; NAPI_GRO_CB(p)->flush = 0; } diff --git a/net/core/dev.c b/net/core/dev.c index 859e30ff044a..63bd20a75929 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3169,16 +3169,18 @@ normal: } EXPORT_SYMBOL(dev_gro_receive); -static gro_result_t +static inline gro_result_t __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { struct sk_buff *p; for (p = napi->gro_list; p; p = p->next) { - NAPI_GRO_CB(p)->same_flow = - (p->dev == skb->dev) && - !compare_ether_header(skb_mac_header(p), + unsigned long diffs; + + diffs = (unsigned long)p->dev ^ (unsigned long)skb->dev; + diffs |= compare_ether_header(skb_mac_header(p), skb_gro_mac_header(skb)); + NAPI_GRO_CB(p)->same_flow = !diffs; NAPI_GRO_CB(p)->flush = 0; } -- cgit v1.2.3 From ba4fd9d8282f7f856f2287fe8be784d1dfdda28b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 31 Aug 2010 01:57:35 +0000 Subject: pktgen: remove non used variable remove non used variable "queue" in pg_cleanup Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/pktgen.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net/core') diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 10a1ea72010d..386c2283f14e 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -3907,8 +3907,6 @@ static void __exit pg_cleanup(void) { struct pktgen_thread *t; struct list_head *q, *n; - wait_queue_head_t queue; - init_waitqueue_head(&queue); /* Stop all interfaces & threads */ -- cgit v1.2.3 From 86cac58b71227cc34a3d0e78f19585c0eff49ea3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 31 Aug 2010 18:25:32 +0000 Subject: skge: add GRO support - napi_gro_flush() is exported from net/core/dev.c, to avoid an irq_save/irq_restore in the packet receive path. - use napi_gro_receive() instead of netif_receive_skb() - use napi_gro_flush() before calling __napi_complete() - turn on NETIF_F_GRO by default - Tested on a Marvell 88E8001 Gigabit NIC Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/skge.c | 5 +++-- include/linux/netdevice.h | 1 + net/core/dev.c | 3 ++- 3 files changed, 6 insertions(+), 3 deletions(-) (limited to 'net/core') diff --git a/drivers/net/skge.c b/drivers/net/skge.c index 40e5c46e7571..a8a63581d63d 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -3178,8 +3178,7 @@ static int skge_poll(struct napi_struct *napi, int to_do) skb = skge_rx_get(dev, e, control, rd->status, rd->csum2); if (likely(skb)) { - netif_receive_skb(skb); - + napi_gro_receive(napi, skb); ++work_done; } } @@ -3192,6 +3191,7 @@ static int skge_poll(struct napi_struct *napi, int to_do) if (work_done < to_do) { unsigned long flags; + napi_gro_flush(napi); spin_lock_irqsave(&hw->hw_lock, flags); __napi_complete(napi); hw->intr_mask |= napimask[skge->port]; @@ -3849,6 +3849,7 @@ static struct net_device *skge_devinit(struct skge_hw *hw, int port, dev->features |= NETIF_F_IP_CSUM | NETIF_F_SG; skge->rx_csum = 1; } + dev->features |= NETIF_F_GRO; /* read the mac address */ memcpy_fromio(dev->dev_addr, hw->regs + B2_MAC_1 + port*8, ETH_ALEN); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c82220a9f3d5..af05186d5b36 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1702,6 +1702,7 @@ extern gro_result_t dev_gro_receive(struct napi_struct *napi, extern gro_result_t napi_skb_finish(gro_result_t ret, struct sk_buff *skb); extern gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb); +extern void napi_gro_flush(struct napi_struct *napi); extern void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb); extern struct sk_buff * napi_get_frags(struct napi_struct *napi); diff --git a/net/core/dev.c b/net/core/dev.c index 63bd20a75929..d8c43e73f0b7 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3063,7 +3063,7 @@ out: return netif_receive_skb(skb); } -static void napi_gro_flush(struct napi_struct *napi) +inline void napi_gro_flush(struct napi_struct *napi) { struct sk_buff *skb, *next; @@ -3076,6 +3076,7 @@ static void napi_gro_flush(struct napi_struct *napi) napi->gro_count = 0; napi->gro_list = NULL; } +EXPORT_SYMBOL(napi_gro_flush); enum gro_result dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { -- cgit v1.2.3 From 6602cebb5bcac1fccf2850541f8bf9fcc8c86dee Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 1 Sep 2010 05:25:10 +0000 Subject: net: skbuff.c cleanup (skb->data - skb->head) can be changed by skb_headroom(skb) Remove some uses of NET_SKBUFF_DATA_USES_OFFSET, using (skb_end_pointer(skb) - skb->head) or (skb_tail_pointer(skb) - skb->head) : compiler does the right thing, and this is more readable for us ;) (struct skb_shared_info *) casts in pskb_expand_head() to help memcpy() to use aligned moves. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/skbuff.c | 47 +++++++++++++++-------------------------------- 1 file changed, 15 insertions(+), 32 deletions(-) (limited to 'net/core') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index e2535fb4985d..231dff0dde2e 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -685,16 +685,10 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old) struct sk_buff *skb_copy(const struct sk_buff *skb, gfp_t gfp_mask) { - int headerlen = skb->data - skb->head; - /* - * Allocate the copy buffer - */ - struct sk_buff *n; -#ifdef NET_SKBUFF_DATA_USES_OFFSET - n = alloc_skb(skb->end + skb->data_len, gfp_mask); -#else - n = alloc_skb(skb->end - skb->head + skb->data_len, gfp_mask); -#endif + int headerlen = skb_headroom(skb); + unsigned int size = (skb_end_pointer(skb) - skb->head) + skb->data_len; + struct sk_buff *n = alloc_skb(size, gfp_mask); + if (!n) return NULL; @@ -726,20 +720,14 @@ EXPORT_SYMBOL(skb_copy); struct sk_buff *pskb_copy(struct sk_buff *skb, gfp_t gfp_mask) { - /* - * Allocate the copy buffer - */ - struct sk_buff *n; -#ifdef NET_SKBUFF_DATA_USES_OFFSET - n = alloc_skb(skb->end, gfp_mask); -#else - n = alloc_skb(skb->end - skb->head, gfp_mask); -#endif + unsigned int size = skb_end_pointer(skb) - skb->head; + struct sk_buff *n = alloc_skb(size, gfp_mask); + if (!n) goto out; /* Set the data pointer */ - skb_reserve(n, skb->data - skb->head); + skb_reserve(n, skb_headroom(skb)); /* Set the tail pointer and length */ skb_put(n, skb_headlen(skb)); /* Copy the bytes */ @@ -791,11 +779,7 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, { int i; u8 *data; -#ifdef NET_SKBUFF_DATA_USES_OFFSET - int size = nhead + skb->end + ntail; -#else - int size = nhead + (skb->end - skb->head) + ntail; -#endif + int size = nhead + (skb_end_pointer(skb) - skb->head) + ntail; long off; BUG_ON(nhead < 0); @@ -810,13 +794,12 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, goto nodata; /* Copy only real data... and, alas, header. This should be - * optimized for the cases when header is void. */ -#ifdef NET_SKBUFF_DATA_USES_OFFSET - memcpy(data + nhead, skb->head, skb->tail); -#else - memcpy(data + nhead, skb->head, skb->tail - skb->head); -#endif - memcpy(data + size, skb_end_pointer(skb), + * optimized for the cases when header is void. + */ + memcpy(data + nhead, skb->head, skb_tail_pointer(skb) - skb->head); + + memcpy((struct skb_shared_info *)(data + size), + skb_shinfo(skb), offsetof(struct skb_shared_info, frags[skb_shinfo(skb)->nr_frags])); for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) -- cgit v1.2.3 From fa50d6457691d5c2d8a3430abf950435ef129cf1 Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Tue, 31 Aug 2010 12:14:13 +0000 Subject: net: make rx_queue sysfs_ops const Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/core') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 7d748542d97e..76485a3f910b 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -515,7 +515,7 @@ static ssize_t rx_queue_attr_store(struct kobject *kobj, struct attribute *attr, return attribute->store(queue, attribute, buf, count); } -static struct sysfs_ops rx_queue_sysfs_ops = { +static const struct sysfs_ops rx_queue_sysfs_ops = { .show = rx_queue_attr_show, .store = rx_queue_attr_store, }; -- cgit v1.2.3 From c07b68e841bd737e2ebeb57268d251c89ccc5010 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 2 Sep 2010 03:53:46 +0000 Subject: net: dev_add_pack() & __dev_remove_pack() changes Add a small helper ptype_head() to get the head to manipulate dev_add_pack() & __dev_remove_pack() can use a spinlock without blocking BH, since softirq use RCU, and these functions are run from process context only. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 32 +++++++++++++++----------------- 1 file changed, 15 insertions(+), 17 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index d8c43e73f0b7..efd318db11ab 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -371,6 +371,14 @@ static inline void netdev_set_addr_lockdep_class(struct net_device *dev) * --ANK (980803) */ +static inline struct list_head *ptype_head(const struct packet_type *pt) +{ + if (pt->type == htons(ETH_P_ALL)) + return &ptype_all; + else + return &ptype_base[ntohs(pt->type) & PTYPE_HASH_MASK]; +} + /** * dev_add_pack - add packet handler * @pt: packet type declaration @@ -386,16 +394,11 @@ static inline void netdev_set_addr_lockdep_class(struct net_device *dev) void dev_add_pack(struct packet_type *pt) { - int hash; + struct list_head *head = ptype_head(pt); - spin_lock_bh(&ptype_lock); - if (pt->type == htons(ETH_P_ALL)) - list_add_rcu(&pt->list, &ptype_all); - else { - hash = ntohs(pt->type) & PTYPE_HASH_MASK; - list_add_rcu(&pt->list, &ptype_base[hash]); - } - spin_unlock_bh(&ptype_lock); + spin_lock(&ptype_lock); + list_add_rcu(&pt->list, head); + spin_unlock(&ptype_lock); } EXPORT_SYMBOL(dev_add_pack); @@ -414,15 +417,10 @@ EXPORT_SYMBOL(dev_add_pack); */ void __dev_remove_pack(struct packet_type *pt) { - struct list_head *head; + struct list_head *head = ptype_head(pt); struct packet_type *pt1; - spin_lock_bh(&ptype_lock); - - if (pt->type == htons(ETH_P_ALL)) - head = &ptype_all; - else - head = &ptype_base[ntohs(pt->type) & PTYPE_HASH_MASK]; + spin_lock(&ptype_lock); list_for_each_entry(pt1, head, list) { if (pt == pt1) { @@ -433,7 +431,7 @@ void __dev_remove_pack(struct packet_type *pt) printk(KERN_WARNING "dev_remove_pack: %p not found.\n", pt); out: - spin_unlock_bh(&ptype_lock); + spin_unlock(&ptype_lock); } EXPORT_SYMBOL(__dev_remove_pack); -- cgit v1.2.3 From 52ee7a04a0f88815a71acdc604a854fb30dcbe45 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 Sep 2010 06:27:08 +0000 Subject: net: remove two kmemcheck annotations __alloc_skb() uses a memset() to clear all the beginning of skb, including bitfields contained in 'flags1' & 'flags2'. We dont need any more to use kmemcheck_annotate_bitfield() on these fields. However, we still need it for the clone part, which is not cleared. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/skbuff.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net/core') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 231dff0dde2e..c030cf894f57 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -202,8 +202,6 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, skb->data = data; skb_reset_tail_pointer(skb); skb->end = skb->tail + size; - kmemcheck_annotate_bitfield(skb, flags1); - kmemcheck_annotate_bitfield(skb, flags2); #ifdef NET_SKBUFF_DATA_USES_OFFSET skb->mac_header = ~0U; #endif -- cgit v1.2.3 From 1fd63041c49c5c6ed1fe58b7bccc2de462d51e2b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 2 Sep 2010 23:09:32 +0000 Subject: net: pskb_expand_head() optimization pskb_expand_head() blindly takes references on fragments before calling skb_release_data(), potentially releasing these references. We can add a fast path, avoiding these atomic operations, if we own the last reference on skb->head. Based on a previous patch from David Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/skbuff.c | 25 ++++++++++++++++++++----- 1 file changed, 20 insertions(+), 5 deletions(-) (limited to 'net/core') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index c030cf894f57..2d1bc761fe4b 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -779,6 +779,7 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, u8 *data; int size = nhead + (skb_end_pointer(skb) - skb->head) + ntail; long off; + bool fastpath; BUG_ON(nhead < 0); @@ -800,14 +801,28 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, skb_shinfo(skb), offsetof(struct skb_shared_info, frags[skb_shinfo(skb)->nr_frags])); - for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) - get_page(skb_shinfo(skb)->frags[i].page); + /* Check if we can avoid taking references on fragments if we own + * the last reference on skb->head. (see skb_release_data()) + */ + if (!skb->cloned) + fastpath = true; + else { + int delta = skb->nohdr ? (1 << SKB_DATAREF_SHIFT) + 1 : 1; - if (skb_has_frag_list(skb)) - skb_clone_fraglist(skb); + fastpath = atomic_read(&skb_shinfo(skb)->dataref) == delta; + } - skb_release_data(skb); + if (fastpath) { + kfree(skb->head); + } else { + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) + get_page(skb_shinfo(skb)->frags[i].page); + if (skb_has_frag_list(skb)) + skb_clone_fraglist(skb); + + skb_release_data(skb); + } off = (data + nhead) - skb->head; skb->head = data; -- cgit v1.2.3 From db40980fcdb560d7992b0511df16cdd3f7e381f3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 6 Sep 2010 11:13:50 +0000 Subject: net: poll() optimizations No need to test twice sk->sk_shutdown & RCV_SHUTDOWN Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/bluetooth/af_bluetooth.c | 5 ++--- net/core/datagram.c | 5 ++--- net/sctp/socket.c | 5 ++--- net/unix/af_unix.c | 5 ++--- 4 files changed, 8 insertions(+), 12 deletions(-) (limited to 'net/core') diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 421c45bd1b95..ed0f22f57668 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -297,13 +297,12 @@ unsigned int bt_sock_poll(struct file * file, struct socket *sock, poll_table *w mask |= POLLERR; if (sk->sk_shutdown & RCV_SHUTDOWN) - mask |= POLLRDHUP; + mask |= POLLRDHUP | POLLIN | POLLRDNORM; if (sk->sk_shutdown == SHUTDOWN_MASK) mask |= POLLHUP; - if (!skb_queue_empty(&sk->sk_receive_queue) || - (sk->sk_shutdown & RCV_SHUTDOWN)) + if (!skb_queue_empty(&sk->sk_receive_queue)) mask |= POLLIN | POLLRDNORM; if (sk->sk_state == BT_CLOSED) diff --git a/net/core/datagram.c b/net/core/datagram.c index 251997a95483..4df1b7a6c1bf 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -746,13 +746,12 @@ unsigned int datagram_poll(struct file *file, struct socket *sock, if (sk->sk_err || !skb_queue_empty(&sk->sk_error_queue)) mask |= POLLERR; if (sk->sk_shutdown & RCV_SHUTDOWN) - mask |= POLLRDHUP; + mask |= POLLRDHUP | POLLIN | POLLRDNORM; if (sk->sk_shutdown == SHUTDOWN_MASK) mask |= POLLHUP; /* readable? */ - if (!skb_queue_empty(&sk->sk_receive_queue) || - (sk->sk_shutdown & RCV_SHUTDOWN)) + if (!skb_queue_empty(&sk->sk_receive_queue)) mask |= POLLIN | POLLRDNORM; /* Connection-based need to check for termination and startup */ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index f4bec2772351..cf6dcc908b88 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5707,13 +5707,12 @@ unsigned int sctp_poll(struct file *file, struct socket *sock, poll_table *wait) if (sk->sk_err || !skb_queue_empty(&sk->sk_error_queue)) mask |= POLLERR; if (sk->sk_shutdown & RCV_SHUTDOWN) - mask |= POLLRDHUP; + mask |= POLLRDHUP | POLLIN | POLLRDNORM; if (sk->sk_shutdown == SHUTDOWN_MASK) mask |= POLLHUP; /* Is it readable? Reconsider this code with TCP-style support. */ - if (!skb_queue_empty(&sk->sk_receive_queue) || - (sk->sk_shutdown & RCV_SHUTDOWN)) + if (!skb_queue_empty(&sk->sk_receive_queue)) mask |= POLLIN | POLLRDNORM; /* The association is either gone or not ready. */ diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 4414a18c63b4..9077b4ea00c5 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2024,11 +2024,10 @@ static unsigned int unix_poll(struct file *file, struct socket *sock, poll_table if (sk->sk_shutdown == SHUTDOWN_MASK) mask |= POLLHUP; if (sk->sk_shutdown & RCV_SHUTDOWN) - mask |= POLLRDHUP; + mask |= POLLRDHUP | POLLIN | POLLRDNORM; /* readable? */ - if (!skb_queue_empty(&sk->sk_receive_queue) || - (sk->sk_shutdown & RCV_SHUTDOWN)) + if (!skb_queue_empty(&sk->sk_receive_queue)) mask |= POLLIN | POLLRDNORM; /* Connection-based need to check for termination and startup */ -- cgit v1.2.3 From 6febfca98f25c7ee5c3ff7fc85e048bf82230ad5 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Fri, 3 Sep 2010 23:12:37 +0000 Subject: net: rps: add the shortcut for one rps_cpus When there is only one rps_cpus, skb_get_rxhash() can be eliminated. Signed-off-by: Changli Gao Signed-off-by: David S. Miller --- net/core/dev.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index efd318db11ab..cdbbea39c549 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2343,7 +2343,7 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, struct rps_dev_flow **rflowp) { struct netdev_rx_queue *rxqueue; - struct rps_map *map; + struct rps_map *map = NULL; struct rps_dev_flow_table *flow_table; struct rps_sock_flow_table *sock_flow_table; int cpu = -1; @@ -2361,8 +2361,17 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, } else rxqueue = dev->_rx; - if (!rxqueue->rps_map && !rxqueue->rps_flow_table) + if (rxqueue->rps_map) { + map = rcu_dereference(rxqueue->rps_map); + if (map && map->len == 1) { + tcpu = map->cpus[0]; + if (cpu_online(tcpu)) + cpu = tcpu; + goto done; + } + } else if (!rxqueue->rps_flow_table) { goto done; + } skb_reset_network_header(skb); if (!skb_get_rxhash(skb)) @@ -2407,7 +2416,6 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, } } - map = rcu_dereference(rxqueue->rps_map); if (map) { tcpu = map->cpus[((u64) skb->rxhash * map->len) >> 32]; -- cgit v1.2.3 From a700d8be733bd593ea4797dfde17aed4f35213c0 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Wed, 8 Sep 2010 03:48:47 +0000 Subject: net/core: remove address space warnings on verify_iovec() move_addr_to_kernel() and copy_from_user() requires their argument as __user pointer but were missing proper markups. Add it. This removes following warnings from sparse. net/core/iovec.c:44:52: warning: incorrect type in argument 1 (different address spaces) net/core/iovec.c:44:52: expected void [noderef] *uaddr net/core/iovec.c:44:52: got void *msg_name net/core/iovec.c:55:34: warning: incorrect type in argument 2 (different address spaces) net/core/iovec.c:55:34: expected void const [noderef] *from net/core/iovec.c:55:34: got struct iovec *msg_iov Signed-off-by: Namhyung Kim Signed-off-by: David S. Miller --- net/core/iovec.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net/core') diff --git a/net/core/iovec.c b/net/core/iovec.c index 1cd98df412df..f4657c2127b4 100644 --- a/net/core/iovec.c +++ b/net/core/iovec.c @@ -41,7 +41,9 @@ int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *address, if (m->msg_namelen) { if (mode == VERIFY_READ) { - err = move_addr_to_kernel(m->msg_name, m->msg_namelen, + void __user *namep; + namep = (void __user __force *) m->msg_name; + err = move_addr_to_kernel(namep, m->msg_namelen, address); if (err < 0) return err; @@ -52,7 +54,7 @@ int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *address, } size = m->msg_iovlen * sizeof(struct iovec); - if (copy_from_user(iov, m->msg_iov, size)) + if (copy_from_user(iov, (void __user __force *) m->msg_iov, size)) return -EFAULT; m->msg_iov = iov; -- cgit v1.2.3 From f39234d60617d37818b30991e6794643ce220296 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Wed, 8 Sep 2010 03:48:48 +0000 Subject: net/core: add lock context change annotations in net/core/sock.c __lock_sock() and __release_sock() releases and regrabs lock but were missing proper annotations. Add it. This removes following warning from sparse. (Currently __lock_sock() does not emit any warning about it but I think it is better to add also.) net/core/sock.c:1580:17: warning: context imbalance in '__release_sock' - unexpected unlock Signed-off-by: Namhyung Kim Signed-off-by: David S. Miller --- net/core/sock.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net/core') diff --git a/net/core/sock.c b/net/core/sock.c index b05b9b6ddb87..f3a06c40d5e0 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1557,6 +1557,8 @@ struct sk_buff *sock_alloc_send_skb(struct sock *sk, unsigned long size, EXPORT_SYMBOL(sock_alloc_send_skb); static void __lock_sock(struct sock *sk) + __releases(&sk->sk_lock.slock) + __acquires(&sk->sk_lock.slock) { DEFINE_WAIT(wait); @@ -1573,6 +1575,8 @@ static void __lock_sock(struct sock *sk) } static void __release_sock(struct sock *sk) + __releases(&sk->sk_lock.slock) + __acquires(&sk->sk_lock.slock) { struct sk_buff *skb = sk->sk_backlog.head; -- cgit v1.2.3 From 9ca7f8762299bb391c11a81c844224216e925b5c Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Wed, 8 Sep 2010 09:16:28 +0000 Subject: pkt_sched: remov unnecessary bh_disable Now that est_tree_lock is acquired with BH protection, the other call is unnecessary. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/core/gen_estimator.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/core') diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c index 6743146e4d6b..7c2373321b74 100644 --- a/net/core/gen_estimator.c +++ b/net/core/gen_estimator.c @@ -274,9 +274,9 @@ void gen_kill_estimator(struct gnet_stats_basic_packed *bstats, while ((e = gen_find_node(bstats, rate_est))) { rb_erase(&e->node, &est_root); - write_lock_bh(&est_lock); + write_lock(&est_lock); e->bstats = NULL; - write_unlock_bh(&est_lock); + write_unlock(&est_lock); list_del_rcu(&e->list); call_rcu(&e->e_rcu, __gen_kill_estimator); -- cgit v1.2.3 From 83b6b1f5d13414d0cb5c4f0a567a6aec0af073bd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 10 Sep 2010 07:00:25 +0000 Subject: flow: better memory management Allocate hash tables for every online cpus, not every possible ones. NUMA aware allocations. Dont use a full page on arches where PAGE_SIZE > 1024*sizeof(void *) misc: __percpu , __read_mostly, __cpuinit annotations flow_compare_t is just an "unsigned long" Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/flow.c | 78 +++++++++++++++++++++++++++++++-------------------------- 1 file changed, 42 insertions(+), 36 deletions(-) (limited to 'net/core') diff --git a/net/core/flow.c b/net/core/flow.c index f67dcbfe54ef..b143b86b1f2a 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -53,8 +53,7 @@ struct flow_flush_info { struct flow_cache { u32 hash_shift; - unsigned long order; - struct flow_cache_percpu *percpu; + struct flow_cache_percpu __percpu *percpu; struct notifier_block hotcpu_notifier; int low_watermark; int high_watermark; @@ -64,7 +63,7 @@ struct flow_cache { atomic_t flow_cache_genid = ATOMIC_INIT(0); EXPORT_SYMBOL(flow_cache_genid); static struct flow_cache flow_cache_global; -static struct kmem_cache *flow_cachep; +static struct kmem_cache *flow_cachep __read_mostly; static DEFINE_SPINLOCK(flow_cache_gc_lock); static LIST_HEAD(flow_cache_gc_list); @@ -181,11 +180,7 @@ static u32 flow_hash_code(struct flow_cache *fc, & (flow_cache_hash_size(fc) - 1)); } -#if (BITS_PER_LONG == 64) -typedef u64 flow_compare_t; -#else -typedef u32 flow_compare_t; -#endif +typedef unsigned long flow_compare_t; /* I hear what you're saying, use memcmp. But memcmp cannot make * important assumptions that we can here, such as alignment and @@ -357,62 +352,73 @@ void flow_cache_flush(void) put_online_cpus(); } -static void __init flow_cache_cpu_prepare(struct flow_cache *fc, - struct flow_cache_percpu *fcp) +static int __cpuinit flow_cache_cpu_prepare(struct flow_cache *fc, int cpu) { - fcp->hash_table = (struct hlist_head *) - __get_free_pages(GFP_KERNEL|__GFP_ZERO, fc->order); - if (!fcp->hash_table) - panic("NET: failed to allocate flow cache order %lu\n", fc->order); + struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, cpu); + size_t sz = sizeof(struct hlist_head) * flow_cache_hash_size(fc); - fcp->hash_rnd_recalc = 1; - fcp->hash_count = 0; - tasklet_init(&fcp->flush_tasklet, flow_cache_flush_tasklet, 0); + if (!fcp->hash_table) { + fcp->hash_table = kzalloc_node(sz, GFP_KERNEL, cpu_to_node(cpu)); + if (!fcp->hash_table) { + pr_err("NET: failed to allocate flow cache sz %zu\n", sz); + return -ENOMEM; + } + fcp->hash_rnd_recalc = 1; + fcp->hash_count = 0; + tasklet_init(&fcp->flush_tasklet, flow_cache_flush_tasklet, 0); + } + return 0; } -static int flow_cache_cpu(struct notifier_block *nfb, +static int __cpuinit flow_cache_cpu(struct notifier_block *nfb, unsigned long action, void *hcpu) { struct flow_cache *fc = container_of(nfb, struct flow_cache, hotcpu_notifier); - int cpu = (unsigned long) hcpu; + int res, cpu = (unsigned long) hcpu; struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, cpu); - if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) + switch (action) { + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + res = flow_cache_cpu_prepare(fc, cpu); + if (res) + return notifier_from_errno(res); + break; + case CPU_DEAD: + case CPU_DEAD_FROZEN: __flow_cache_shrink(fc, fcp, 0); + break; + } return NOTIFY_OK; } -static int flow_cache_init(struct flow_cache *fc) +static int __init flow_cache_init(struct flow_cache *fc) { - unsigned long order; int i; fc->hash_shift = 10; fc->low_watermark = 2 * flow_cache_hash_size(fc); fc->high_watermark = 4 * flow_cache_hash_size(fc); - for (order = 0; - (PAGE_SIZE << order) < - (sizeof(struct hlist_head)*flow_cache_hash_size(fc)); - order++) - /* NOTHING */; - fc->order = order; fc->percpu = alloc_percpu(struct flow_cache_percpu); + if (!fc->percpu) + return -ENOMEM; - setup_timer(&fc->rnd_timer, flow_cache_new_hashrnd, - (unsigned long) fc); - fc->rnd_timer.expires = jiffies + FLOW_HASH_RND_PERIOD; - add_timer(&fc->rnd_timer); - - for_each_possible_cpu(i) - flow_cache_cpu_prepare(fc, per_cpu_ptr(fc->percpu, i)); - + for_each_online_cpu(i) { + if (flow_cache_cpu_prepare(fc, i)) + return -ENOMEM; + } fc->hotcpu_notifier = (struct notifier_block){ .notifier_call = flow_cache_cpu, }; register_hotcpu_notifier(&fc->hotcpu_notifier); + setup_timer(&fc->rnd_timer, flow_cache_new_hashrnd, + (unsigned long) fc); + fc->rnd_timer.expires = jiffies + FLOW_HASH_RND_PERIOD; + add_timer(&fc->rnd_timer); + return 0; } -- cgit v1.2.3 From e0de7c93b950b9e784894efc4b529c6958cb747a Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Tue, 14 Sep 2010 09:13:08 +0000 Subject: ethtool: Remove unimplemented flow specification types struct ethtool_rawip4_spec and struct ethtool_ether_spec are neither commented nor used by any driver, so remove them. Adjust padding in the user-visible unions that included these structures. Fix references to struct ethtool_rawip4_spec in ethtool_get_rx_ntuple(), which should use struct ethtool_usrip4_spec. struct ethtool_usrip4_spec cannot hold IPv6 host addresses and there is no separate structure that can, so remove ETH_RX_NFC_IP6 and the reference to it in niu. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- drivers/net/niu.c | 19 +++++-------------- include/linux/ethtool.h | 21 ++------------------- net/core/ethtool.c | 8 ++++---- 3 files changed, 11 insertions(+), 37 deletions(-) (limited to 'net/core') diff --git a/drivers/net/niu.c b/drivers/net/niu.c index 8e1859c801a4..e36a83845a1c 100644 --- a/drivers/net/niu.c +++ b/drivers/net/niu.c @@ -7462,10 +7462,12 @@ static int niu_add_ethtool_tcam_entry(struct niu *np, if (fsp->flow_type == IP_USER_FLOW) { int i; int add_usr_cls = 0; - int ipv6 = 0; struct ethtool_usrip4_spec *uspec = &fsp->h_u.usr_ip4_spec; struct ethtool_usrip4_spec *umask = &fsp->m_u.usr_ip4_spec; + if (uspec->ip_ver != ETH_RX_NFC_IP4) + return -EINVAL; + niu_lock_parent(np, flags); for (i = 0; i < NIU_L3_PROG_CLS; i++) { @@ -7494,9 +7496,7 @@ static int niu_add_ethtool_tcam_entry(struct niu *np, default: break; } - if (uspec->ip_ver == ETH_RX_NFC_IP6) - ipv6 = 1; - ret = tcam_user_ip_class_set(np, class, ipv6, + ret = tcam_user_ip_class_set(np, class, 0, uspec->proto, uspec->tos, umask->tos); @@ -7553,16 +7553,7 @@ static int niu_add_ethtool_tcam_entry(struct niu *np, ret = -EINVAL; goto out; case IP_USER_FLOW: - if (fsp->h_u.usr_ip4_spec.ip_ver == ETH_RX_NFC_IP4) { - niu_get_tcamkey_from_ip4fs(fsp, tp, l2_rdc_table, - class); - } else { - /* Not yet implemented */ - netdev_info(np->dev, "niu%d: In %s(): usr flow for IPv6 not implemented\n", - parent->index, __func__); - ret = -EINVAL; - goto out; - } + niu_get_tcamkey_from_ip4fs(fsp, tp, l2_rdc_table, class); break; default: netdev_info(np->dev, "niu%d: In %s(): Unknown flow type %d\n", diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index 4b3ba05b11a8..d64e246a39e7 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -353,20 +353,7 @@ struct ethtool_ah_espip4_spec { __u8 tos; }; -struct ethtool_rawip4_spec { - __be32 ip4src; - __be32 ip4dst; - __u8 hdata[64]; -}; - -struct ethtool_ether_spec { - __be16 ether_type; - __u8 frame_size; - __u8 eframe[16]; -}; - #define ETH_RX_NFC_IP4 1 -#define ETH_RX_NFC_IP6 2 /** * struct ethtool_usrip4_spec - general flow specification for IPv4 @@ -403,10 +390,8 @@ struct ethtool_rx_flow_spec { struct ethtool_tcpip4_spec sctp_ip4_spec; struct ethtool_ah_espip4_spec ah_ip4_spec; struct ethtool_ah_espip4_spec esp_ip4_spec; - struct ethtool_rawip4_spec raw_ip4_spec; - struct ethtool_ether_spec ether_spec; struct ethtool_usrip4_spec usr_ip4_spec; - __u8 hdata[64]; + __u8 hdata[72]; } h_u, m_u; __u64 ring_cookie; __u32 location; @@ -496,10 +481,8 @@ struct ethtool_rx_ntuple_flow_spec { struct ethtool_tcpip4_spec sctp_ip4_spec; struct ethtool_ah_espip4_spec ah_ip4_spec; struct ethtool_ah_espip4_spec esp_ip4_spec; - struct ethtool_rawip4_spec raw_ip4_spec; - struct ethtool_ether_spec ether_spec; struct ethtool_usrip4_spec usr_ip4_spec; - __u8 hdata[64]; + __u8 hdata[72]; } h_u, m_u; __u16 vlan_tag; diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 970eb9817bbc..fcd62757704d 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -673,19 +673,19 @@ static int ethtool_get_rx_ntuple(struct net_device *dev, void __user *useraddr) break; case IP_USER_FLOW: sprintf(p, "\tSrc IP addr: 0x%x\n", - fsc->fs.h_u.raw_ip4_spec.ip4src); + fsc->fs.h_u.usr_ip4_spec.ip4src); p += ETH_GSTRING_LEN; num_strings++; sprintf(p, "\tSrc IP mask: 0x%x\n", - fsc->fs.m_u.raw_ip4_spec.ip4src); + fsc->fs.m_u.usr_ip4_spec.ip4src); p += ETH_GSTRING_LEN; num_strings++; sprintf(p, "\tDest IP addr: 0x%x\n", - fsc->fs.h_u.raw_ip4_spec.ip4dst); + fsc->fs.h_u.usr_ip4_spec.ip4dst); p += ETH_GSTRING_LEN; num_strings++; sprintf(p, "\tDest IP mask: 0x%x\n", - fsc->fs.m_u.raw_ip4_spec.ip4dst); + fsc->fs.m_u.usr_ip4_spec.ip4dst); p += ETH_GSTRING_LEN; num_strings++; break; -- cgit v1.2.3 From 95ae6b228f814fc0528d0506ee9f18ac333d6851 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 15 Sep 2010 04:04:31 +0000 Subject: ipv4: ip_ptr cleanups dev->ip_ptr is protected by rtnl and rcu. Yet some places dont use appropriate primitives and/or locking rules. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/plip.c | 8 ++++++-- drivers/net/via-velocity.h | 11 +++++++---- drivers/net/wan/hdlc_cisco.c | 4 +++- include/linux/inetdevice.h | 14 +++++--------- include/linux/netdevice.h | 2 +- net/core/dev.c | 2 +- net/ipv4/devinet.c | 4 ++-- net/ipv4/ipmr.c | 2 +- net/mac80211/main.c | 2 +- 9 files changed, 27 insertions(+), 22 deletions(-) (limited to 'net/core') diff --git a/drivers/net/plip.c b/drivers/net/plip.c index 7e82a82422cf..ca4df7f4cf21 100644 --- a/drivers/net/plip.c +++ b/drivers/net/plip.c @@ -995,8 +995,10 @@ plip_tx_packet(struct sk_buff *skb, struct net_device *dev) static void plip_rewrite_address(const struct net_device *dev, struct ethhdr *eth) { - const struct in_device *in_dev = dev->ip_ptr; + const struct in_device *in_dev; + rcu_read_lock(); + in_dev = __in_dev_get_rcu(dev); if (in_dev) { /* Any address will do - we take the first */ const struct in_ifaddr *ifa = in_dev->ifa_list; @@ -1006,6 +1008,7 @@ plip_rewrite_address(const struct net_device *dev, struct ethhdr *eth) memcpy(eth->h_dest+2, &ifa->ifa_address, 4); } } + rcu_read_unlock(); } static int @@ -1088,7 +1091,8 @@ plip_open(struct net_device *dev) when the device address isn't identical to the address of a received frame, the kernel incorrectly drops it). */ - if ((in_dev=dev->ip_ptr) != NULL) { + in_dev=__in_dev_get_rtnl(dev); + if (in_dev) { /* Any address will do - we take the first. We already have the first two bytes filled with 0xfc, from plip_init_dev(). */ diff --git a/drivers/net/via-velocity.h b/drivers/net/via-velocity.h index f7b33ae7a703..b5e120b0074b 100644 --- a/drivers/net/via-velocity.h +++ b/drivers/net/via-velocity.h @@ -1504,22 +1504,25 @@ struct velocity_info { * addresses on this chain then we use the first - multi-IP WOL is not * supported. * - * CHECK ME: locking */ static inline int velocity_get_ip(struct velocity_info *vptr) { - struct in_device *in_dev = (struct in_device *) vptr->dev->ip_ptr; + struct in_device *in_dev; struct in_ifaddr *ifa; + int res = -ENOENT; + rcu_read_lock(); + in_dev = __in_dev_get_rcu(vptr->dev); if (in_dev != NULL) { ifa = (struct in_ifaddr *) in_dev->ifa_list; if (ifa != NULL) { memcpy(vptr->ip_addr, &ifa->ifa_address, 4); - return 0; + res = 0; } } - return -ENOENT; + rcu_read_unlock(); + return res; } /** diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c index b38ffa149aba..b1e5e5b69c2a 100644 --- a/drivers/net/wan/hdlc_cisco.c +++ b/drivers/net/wan/hdlc_cisco.c @@ -191,7 +191,8 @@ static int cisco_rx(struct sk_buff *skb) switch (ntohl (cisco_data->type)) { case CISCO_ADDR_REQ: /* Stolen from syncppp.c :-) */ - in_dev = dev->ip_ptr; + rcu_read_lock(); + in_dev = __in_dev_get_rcu(dev); addr = 0; mask = ~cpu_to_be32(0); /* is the mask correct? */ @@ -211,6 +212,7 @@ static int cisco_rx(struct sk_buff *skb) cisco_keepalive_send(dev, CISCO_ADDR_REPLY, addr, mask); } + rcu_read_unlock(); dev_kfree_skb_any(skb); return NET_RX_SUCCESS; diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index 2be1a1a2beb9..1ec09bb4a3ab 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -9,6 +9,7 @@ #include #include #include +#include enum { @@ -198,14 +199,10 @@ static __inline__ int bad_mask(__be32 mask, __be32 addr) static inline struct in_device *__in_dev_get_rcu(const struct net_device *dev) { - struct in_device *in_dev = dev->ip_ptr; - if (in_dev) - in_dev = rcu_dereference(in_dev); - return in_dev; + return rcu_dereference(dev->ip_ptr); } -static __inline__ struct in_device * -in_dev_get(const struct net_device *dev) +static inline struct in_device *in_dev_get(const struct net_device *dev) { struct in_device *in_dev; @@ -217,10 +214,9 @@ in_dev_get(const struct net_device *dev) return in_dev; } -static __inline__ struct in_device * -__in_dev_get_rtnl(const struct net_device *dev) +static inline struct in_device *__in_dev_get_rtnl(const struct net_device *dev) { - return (struct in_device*)dev->ip_ptr; + return rcu_dereference_check(dev->ip_ptr, lockdep_rtnl_is_held()); } extern void in_dev_finish_destroy(struct in_device *idev); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index af05186d5b36..8992fffb8104 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -942,7 +942,7 @@ struct net_device { void *dsa_ptr; /* dsa specific data */ #endif void *atalk_ptr; /* AppleTalk link */ - void *ip_ptr; /* IPv4 specific data */ + struct in_device __rcu *ip_ptr; /* IPv4 specific data */ void *dn_ptr; /* DECnet specific data */ void *ip6_ptr; /* IPv6 specific data */ void *ec_ptr; /* Econet specific data */ diff --git a/net/core/dev.c b/net/core/dev.c index fc2dc933bee5..5bdce97b8175 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5286,7 +5286,7 @@ void netdev_run_todo(void) /* paranoia */ BUG_ON(atomic_read(&dev->refcnt)); - WARN_ON(dev->ip_ptr); + WARN_ON(rcu_dereference_raw(dev->ip_ptr)); WARN_ON(dev->ip6_ptr); WARN_ON(dev->dn_ptr); diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index da14c49284f4..c2ff48fa18c7 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -209,7 +209,7 @@ static void inetdev_destroy(struct in_device *in_dev) inet_free_ifa(ifa); } - dev->ip_ptr = NULL; + rcu_assign_pointer(dev->ip_ptr, NULL); devinet_sysctl_unregister(in_dev); neigh_parms_release(&arp_tbl, in_dev->arp_parms); @@ -1059,7 +1059,7 @@ static int inetdev_event(struct notifier_block *this, unsigned long event, switch (event) { case NETDEV_REGISTER: printk(KERN_DEBUG "inetdev_event: bug\n"); - dev->ip_ptr = NULL; + rcu_assign_pointer(dev->ip_ptr, NULL); break; case NETDEV_UP: if (!inetdev_valid_mtu(dev->mtu)) diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 179fcab866fc..10b24c02deb0 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -724,7 +724,7 @@ static int vif_add(struct net *net, struct mr_table *mrt, case 0: if (vifc->vifc_flags == VIFF_USE_IFINDEX) { dev = dev_get_by_index(net, vifc->vifc_lcl_ifindex); - if (dev && dev->ip_ptr == NULL) { + if (dev && __in_dev_get_rtnl(dev) == NULL) { dev_put(dev); return -EADDRNOTAVAIL; } diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 4935b843bcca..b8cf2821f00d 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -362,7 +362,7 @@ static int ieee80211_ifa_changed(struct notifier_block *nb, if (sdata->vif.type != NL80211_IFTYPE_STATION) return NOTIFY_DONE; - idev = sdata->dev->ip_ptr; + idev = __in_dev_get_rtnl(sdata->dev); if (!idev) return NOTIFY_DONE; -- cgit v1.2.3 From 16c3ea785fe4a383c6675dfe7316f3c815755bdd Mon Sep 17 00:00:00 2001 From: Brandon Philips Date: Wed, 15 Sep 2010 09:24:24 +0000 Subject: net: enable GRO by default for vlan devices Currently vlan devices don't have GRO by default as none of the Ethernet drivers add NETIF_F_GRO to their vlan_features. As GRO is a software feature add GRO to dev->vlan_features in register_netdevice() and let vlan_dev_init() take care that it gets enabled only when dev->features has NETIF_F_GRO too. Signed-off-by: Brandon Philips Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 5bdce97b8175..09b3742c4c89 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5057,6 +5057,11 @@ int register_netdevice(struct net_device *dev) if (dev->features & NETIF_F_SG) dev->features |= NETIF_F_GSO; + /* Enable GRO for vlans by default if dev->features has GRO also. + * vlan_dev_init() will do the dev->features check. + */ + dev->vlan_features |= NETIF_F_GRO; + ret = call_netdevice_notifiers(NETDEV_POST_INIT, dev); ret = notifier_to_errno(ret); if (ret) -- cgit v1.2.3 From caeda9b926c608702c99f3432aae2c24298c3c1d Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Thu, 16 Sep 2010 21:39:16 -0700 Subject: net: include inetdevice.h for rcu_dereference_raw api change rcu_dereference_raw() now needs to know the type of its argument. Signed-off-by: Stephen Rothwell Signed-off-by: David S. Miller --- net/core/dev.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 09b3742c4c89..c09ff096525a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -129,6 +129,7 @@ #include #include #include +#include #include "net-sysfs.h" -- cgit v1.2.3 From 67c9660831f6b6b76866a0838466c83765ffbbd3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 17 Sep 2010 11:56:18 -0700 Subject: ethtool: change ethtool_set_gro() to use ethtool_op_get_rx_csum To be able to switch on GRO on a device, ethtool_set_gro() checks this device provides a get_rx_csum() method. Some devices dont provide this method, while they do support RX checksumming. This patch allows bonding to support GRO : ethtool -K bond0 gro on Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/ethtool.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'net/core') diff --git a/net/core/ethtool.c b/net/core/ethtool.c index fcd62757704d..248c25c3e820 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -1181,8 +1181,11 @@ static int ethtool_set_gro(struct net_device *dev, char __user *useraddr) return -EFAULT; if (edata.data) { - if (!dev->ethtool_ops->get_rx_csum || - !dev->ethtool_ops->get_rx_csum(dev)) + u32 rxcsum = dev->ethtool_ops->get_rx_csum ? + dev->ethtool_ops->get_rx_csum(dev) : + ethtool_op_get_rx_csum(dev); + + if (!rxcsum) return -EINVAL; dev->features |= NETIF_F_GRO; } else -- cgit v1.2.3 From 3b27e105550f7c4a79ecb6d6a9c49c651c59ae9b Mon Sep 17 00:00:00 2001 From: David Lamparter Date: Fri, 17 Sep 2010 03:22:19 +0000 Subject: netns: keep vlan slaves on master netns move previously, if a vlan master device was moved from one network namespace to another, all 802.1q and macvlan slaves were deleted. we can use dev->reg_state to figure out whether dev_change_net_namespace is happening, since that won't set dev->reg_state NETREG_UNREGISTERING. so, this changes 8021q and macvlan to ignore NETDEV_UNREGISTER when reg_state is not NETREG_UNREGISTERING. Signed-off-by: David Lamparter Reviewed-by: "Eric W. Biederman" Acked-by: Daniel Lezcano Acked-by: Patrick McHardy Signed-off-by: David S. Miller --- drivers/net/macvlan.c | 4 ++++ net/8021q/vlan.c | 4 ++++ net/core/dev.c | 4 ++++ 3 files changed, 12 insertions(+) (limited to 'net/core') diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 0ef0eb0db945..0fc9dc7f20db 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -788,6 +788,10 @@ static int macvlan_device_event(struct notifier_block *unused, } break; case NETDEV_UNREGISTER: + /* twiddle thumbs on netns device moves */ + if (dev->reg_state != NETREG_UNREGISTERING) + break; + list_for_each_entry_safe(vlan, next, &port->vlans, list) vlan->dev->rtnl_link_ops->dellink(vlan->dev, NULL); break; diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index a2ad15250575..2c6c2bd6e4a9 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -525,6 +525,10 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, break; case NETDEV_UNREGISTER: + /* twiddle thumbs on netns device moves */ + if (dev->reg_state != NETREG_UNREGISTERING) + break; + /* Delete all VLANs for this dev. */ grp->killall = 1; diff --git a/net/core/dev.c b/net/core/dev.c index c09ff096525a..2c7934f8cf3e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5686,6 +5686,10 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char /* Notify protocols, that we are about to destroy this device. They should clean all the things. + + Note that dev->reg_state stays at NETREG_REGISTERED. + This is wanted because this way 8021q and macvlan know + the device is just moving and can keep their slaves up. */ call_netdevice_notifiers(NETDEV_UNREGISTER, dev); call_netdevice_notifiers(NETDEV_UNREGISTER_BATCH, dev); -- cgit v1.2.3 From be2902daee80b655cebd482b5ee91ffc29408121 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 16 Sep 2010 11:28:07 +0000 Subject: ethtool, ixgbe: Move RX n-tuple mask fixup to ethtool The ethtool utility does not set masks for flow parameters that are not specified, so if both value and mask are 0 then this must be treated as equivalent to a mask with all bits set. Currently that is done in the only driver that implements RX n-tuple filtering, ixgbe. Move it to the ethtool core. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- drivers/net/ixgbe/ixgbe_82599.c | 57 ++++++++++------------------------------- include/linux/ethtool.h | 5 ++-- net/core/ethtool.c | 34 ++++++++++++++++++++++++ 3 files changed, 51 insertions(+), 45 deletions(-) (limited to 'net/core') diff --git a/drivers/net/ixgbe/ixgbe_82599.c b/drivers/net/ixgbe/ixgbe_82599.c index 3e06a61da921..e80657c75506 100644 --- a/drivers/net/ixgbe/ixgbe_82599.c +++ b/drivers/net/ixgbe/ixgbe_82599.c @@ -1910,56 +1910,27 @@ s32 ixgbe_fdir_add_perfect_filter_82599(struct ixgbe_hw *hw, (dst_port << IXGBE_FDIRPORT_DESTINATION_SHIFT))); /* - * Program the relevant mask registers. If src/dst_port or src/dst_addr - * are zero, then assume a full mask for that field. Also assume that - * a VLAN of 0 is unspecified, so mask that out as well. L4type - * cannot be masked out in this implementation. + * Program the relevant mask registers. L4type cannot be + * masked out in this implementation. * * This also assumes IPv4 only. IPv6 masking isn't supported at this * point in time. */ - if (src_ipv4 == 0) - IXGBE_WRITE_REG(hw, IXGBE_FDIRSIP4M, 0xffffffff); - else - IXGBE_WRITE_REG(hw, IXGBE_FDIRSIP4M, input_masks->src_ip_mask); - - if (dst_ipv4 == 0) - IXGBE_WRITE_REG(hw, IXGBE_FDIRDIP4M, 0xffffffff); - else - IXGBE_WRITE_REG(hw, IXGBE_FDIRDIP4M, input_masks->dst_ip_mask); + IXGBE_WRITE_REG(hw, IXGBE_FDIRSIP4M, input_masks->src_ip_mask); + IXGBE_WRITE_REG(hw, IXGBE_FDIRDIP4M, input_masks->dst_ip_mask); switch (l4type & IXGBE_ATR_L4TYPE_MASK) { case IXGBE_ATR_L4TYPE_TCP: - if (src_port == 0) - IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM, 0xffff); - else - IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM, - input_masks->src_port_mask); - - if (dst_port == 0) - IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM, - (IXGBE_READ_REG(hw, IXGBE_FDIRTCPM) | - (0xffff << 16))); - else - IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM, - (IXGBE_READ_REG(hw, IXGBE_FDIRTCPM) | - (input_masks->dst_port_mask << 16))); + IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM, input_masks->src_port_mask); + IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM, + (IXGBE_READ_REG(hw, IXGBE_FDIRTCPM) | + (input_masks->dst_port_mask << 16))); break; case IXGBE_ATR_L4TYPE_UDP: - if (src_port == 0) - IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM, 0xffff); - else - IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM, - input_masks->src_port_mask); - - if (dst_port == 0) - IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM, - (IXGBE_READ_REG(hw, IXGBE_FDIRUDPM) | - (0xffff << 16))); - else - IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM, - (IXGBE_READ_REG(hw, IXGBE_FDIRUDPM) | - (input_masks->src_port_mask << 16))); + IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM, input_masks->src_port_mask); + IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM, + (IXGBE_READ_REG(hw, IXGBE_FDIRUDPM) | + (input_masks->src_port_mask << 16))); break; default: /* this already would have failed above */ @@ -1967,11 +1938,11 @@ s32 ixgbe_fdir_add_perfect_filter_82599(struct ixgbe_hw *hw, } /* Program the last mask register, FDIRM */ - if (input_masks->vlan_id_mask || !vlan_id) + if (input_masks->vlan_id_mask) /* Mask both VLAN and VLANP - bits 0 and 1 */ fdirm |= 0x3; - if (input_masks->data_mask || !flex_bytes) + if (input_masks->data_mask) /* Flex bytes need masking, so mask the whole thing - bit 4 */ fdirm |= 0x10; diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index d64e246a39e7..00334eebbe26 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -470,8 +470,9 @@ struct ethtool_rxfh_indir { * @action: RX ring/queue index to deliver to (non-negative) or other action * (negative, e.g. %ETHTOOL_RXNTUPLE_ACTION_DROP) * - * Zero values in @h_u may be ignored, as if all the corresponding - * mask bits were set. + * For flow types %TCP_V4_FLOW, %UDP_V4_FLOW and %SCTP_V4_FLOW, where + * a field value and mask are both zero this is treated as if all mask + * bits are set i.e. the field is ignored. */ struct ethtool_rx_ntuple_flow_spec { __u32 flow_type; diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 248c25c3e820..91ffce20c36b 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -485,6 +485,38 @@ static void __rx_ntuple_filter_add(struct ethtool_rx_ntuple_list *list, list->count++; } +/* + * ethtool does not (or did not) set masks for flow parameters that are + * not specified, so if both value and mask are 0 then this must be + * treated as equivalent to a mask with all bits set. Implement that + * here rather than in drivers. + */ +static void rx_ntuple_fix_masks(struct ethtool_rx_ntuple_flow_spec *fs) +{ + struct ethtool_tcpip4_spec *entry = &fs->h_u.tcp_ip4_spec; + struct ethtool_tcpip4_spec *mask = &fs->m_u.tcp_ip4_spec; + + if (fs->flow_type != TCP_V4_FLOW && + fs->flow_type != UDP_V4_FLOW && + fs->flow_type != SCTP_V4_FLOW) + return; + + if (!(entry->ip4src | mask->ip4src)) + mask->ip4src = htonl(0xffffffff); + if (!(entry->ip4dst | mask->ip4dst)) + mask->ip4dst = htonl(0xffffffff); + if (!(entry->psrc | mask->psrc)) + mask->psrc = htons(0xffff); + if (!(entry->pdst | mask->pdst)) + mask->pdst = htons(0xffff); + if (!(entry->tos | mask->tos)) + mask->tos = 0xff; + if (!(fs->vlan_tag | fs->vlan_tag_mask)) + fs->vlan_tag_mask = 0xffff; + if (!(fs->data | fs->data_mask)) + fs->data_mask = 0xffffffffffffffffULL; +} + static noinline_for_stack int ethtool_set_rx_ntuple(struct net_device *dev, void __user *useraddr) { @@ -499,6 +531,8 @@ static noinline_for_stack int ethtool_set_rx_ntuple(struct net_device *dev, if (copy_from_user(&cmd, useraddr, sizeof(cmd))) return -EFAULT; + rx_ntuple_fix_masks(&cmd.fs); + /* * Cache filter in dev struct for GET operation only if * the underlying driver doesn't have its own GET operation, and -- cgit v1.2.3 From a77f5db361ed9953b5b749353ea2c7fed2bf8d93 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Mon, 20 Sep 2010 08:42:17 +0000 Subject: ethtool: Allocate register dump buffer with vmalloc() Some NICs have huge register files which exceed the maximum heap allocation size. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- net/core/ethtool.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/core') diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 91ffce20c36b..dae2fd053c2b 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -815,7 +815,7 @@ static int ethtool_get_regs(struct net_device *dev, char __user *useraddr) if (regs.len > reglen) regs.len = reglen; - regbuf = kmalloc(reglen, GFP_USER); + regbuf = vmalloc(reglen); if (!regbuf) return -ENOMEM; @@ -830,7 +830,7 @@ static int ethtool_get_regs(struct net_device *dev, char __user *useraddr) ret = 0; out: - kfree(regbuf); + vfree(regbuf); return ret; } -- cgit v1.2.3 From 73da16c28ed0724755afdb95c7dd6b166381be10 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 21 Sep 2010 16:12:11 -0700 Subject: ethtool: Fix build due to lack of ethtool.h include. net/core/ethtool.c: In function 'ethtool_get_regs': net/core/ethtool.c:818:2: error: implicit declaration of function 'vmalloc' net/core/ethtool.c:818:9: warning: assignment makes pointer from integer without a cast net/core/ethtool.c:833:2: error: implicit declaration of function 'vfree' Signed-off-by: David S. Miller --- net/core/ethtool.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/core') diff --git a/net/core/ethtool.c b/net/core/ethtool.c index dae2fd053c2b..7d7e572cedc7 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -19,6 +19,7 @@ #include #include #include +#include #include /* -- cgit v1.2.3 From 82fd5b5d1ec370a50b3060418cde6a4ac8401117 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 20 Sep 2010 20:40:26 +0000 Subject: net: core: use kernel's converter from hex to bin Signed-off-by: Andy Shevchenko Signed-off-by: David S. Miller --- net/core/pktgen.c | 10 ++++------ net/core/utils.c | 13 +++++++------ 2 files changed, 11 insertions(+), 12 deletions(-) (limited to 'net/core') diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 386c2283f14e..2c0df0f95b3d 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -729,16 +729,14 @@ static int hex32_arg(const char __user *user_buffer, unsigned long maxlen, *num = 0; for (; i < maxlen; i++) { + int value; char c; *num <<= 4; if (get_user(c, &user_buffer[i])) return -EFAULT; - if ((c >= '0') && (c <= '9')) - *num |= c - '0'; - else if ((c >= 'a') && (c <= 'f')) - *num |= c - 'a' + 10; - else if ((c >= 'A') && (c <= 'F')) - *num |= c - 'A' + 10; + value = hex_to_bin(c); + if (value >= 0) + *num |= value; else break; } diff --git a/net/core/utils.c b/net/core/utils.c index f41854470539..ec6bb322f372 100644 --- a/net/core/utils.c +++ b/net/core/utils.c @@ -92,18 +92,19 @@ EXPORT_SYMBOL(in_aton); static inline int xdigit2bin(char c, int delim) { + int val; + if (c == delim || c == '\0') return IN6PTON_DELIM; if (c == ':') return IN6PTON_COLON_MASK; if (c == '.') return IN6PTON_DOT; - if (c >= '0' && c <= '9') - return (IN6PTON_XDIGIT | IN6PTON_DIGIT| (c - '0')); - if (c >= 'a' && c <= 'f') - return (IN6PTON_XDIGIT | (c - 'a' + 10)); - if (c >= 'A' && c <= 'F') - return (IN6PTON_XDIGIT | (c - 'A' + 10)); + + val = hex_to_bin(c); + if (val >= 0) + return val | IN6PTON_XDIGIT | (val < 10 ? IN6PTON_DIGIT : 0); + if (delim == -1) return IN6PTON_DELIM; return IN6PTON_UNKNOWN; -- cgit v1.2.3 From a02cec2155fbea457eca8881870fd2de1a4c4c76 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 22 Sep 2010 20:43:57 +0000 Subject: net: return operator cleanup Change "return (EXPR);" to "return EXPR;" return is not a function, parentheses are not required. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/atmdev.h | 2 +- include/linux/etherdevice.h | 4 +-- include/linux/netdevice.h | 2 +- include/linux/skbuff.h | 6 ++--- include/net/bluetooth/hci_core.h | 2 +- include/net/bluetooth/l2cap.h | 2 +- include/net/inet_ecn.h | 2 +- include/net/ip.h | 4 +-- include/net/ipv6.h | 35 +++++++++++++------------- include/net/irda/irlap.h | 2 +- include/net/irda/irlmp.h | 2 +- include/net/irda/irttp.h | 2 +- include/net/sch_generic.h | 2 +- include/net/sctp/sctp.h | 12 ++++----- include/net/sctp/sm.h | 10 ++++---- include/net/sctp/structs.h | 2 +- include/net/sctp/tsnmap.h | 2 +- include/net/tipc/tipc_msg.h | 10 ++++---- net/802/fc.c | 2 +- net/802/fddi.c | 12 ++++----- net/802/hippi.c | 2 +- net/802/tr.c | 2 +- net/8021q/vlan_core.c | 2 +- net/9p/client.c | 4 +-- net/bluetooth/rfcomm/core.c | 4 +-- net/core/flow.c | 4 +-- net/core/neighbour.c | 6 ++--- net/core/utils.c | 2 +- net/dccp/ccids/lib/loss_interval.c | 2 +- net/econet/af_econet.c | 4 +-- net/ethernet/eth.c | 2 +- net/ipv4/arp.c | 2 +- net/ipv4/datagram.c | 2 +- net/ipv4/inet_diag.c | 2 +- net/ipv4/ip_fragment.c | 4 +-- net/ipv4/ip_gre.c | 2 +- net/ipv4/netfilter/arp_tables.c | 2 +- net/ipv4/route.c | 2 +- net/ipv4/tcp_input.c | 10 ++++---- net/ipv4/tcp_minisocks.c | 2 +- net/ipv4/tcp_output.c | 8 +++--- net/ipv4/tcp_westwood.c | 2 +- net/ipv6/addrconf.c | 2 +- net/ipv6/addrlabel.c | 5 ++-- net/ipv6/af_inet6.c | 6 ++--- net/ipv6/exthdrs_core.c | 4 +-- net/ipv6/ip6_output.c | 4 +-- net/ipv6/ndisc.c | 8 +++--- net/ipv6/netfilter/ip6_tables.c | 14 +++++------ net/ipv6/raw.c | 12 ++++----- net/ipv6/route.c | 14 +++++------ net/ipv6/tcp_ipv6.c | 2 +- net/ipv6/xfrm6_policy.c | 2 +- net/irda/af_irda.c | 14 +++++------ net/irda/discovery.c | 2 +- net/irda/ircomm/ircomm_tty.c | 4 +-- net/irda/irlmp.c | 2 +- net/irda/irlmp_frame.c | 2 +- net/irda/irnet/irnet_irda.c | 22 ++++++++--------- net/irda/irnet/irnet_ppp.c | 8 +++--- net/key/af_key.c | 4 +-- net/mac80211/rate.c | 2 +- net/rfkill/input.c | 2 +- net/rose/rose_link.c | 4 +-- net/sctp/protocol.c | 2 +- net/sctp/socket.c | 6 ++--- net/sunrpc/auth_gss/auth_gss.c | 2 +- net/sunrpc/auth_gss/gss_generic_token.c | 44 ++++++++++++++++----------------- net/sunrpc/auth_gss/gss_krb5_seqnum.c | 2 +- net/sunrpc/auth_gss/gss_mech_switch.c | 2 +- net/sunrpc/sched.c | 2 +- net/tipc/addr.c | 2 +- net/tipc/bcast.c | 2 +- net/tipc/bearer.c | 2 +- net/tipc/dbg.c | 4 +-- net/tipc/link.c | 6 ++--- net/tipc/link.h | 16 ++++++------ net/tipc/msg.h | 6 ++--- net/tipc/name_table.c | 2 +- net/tipc/node.c | 6 ++--- net/tipc/port.h | 2 +- net/tipc/socket.c | 2 +- net/tipc/subscr.c | 2 +- net/wireless/core.h | 2 +- 84 files changed, 220 insertions(+), 222 deletions(-) (limited to 'net/core') diff --git a/include/linux/atmdev.h b/include/linux/atmdev.h index f6481daf6e52..a8e4e832cdbb 100644 --- a/include/linux/atmdev.h +++ b/include/linux/atmdev.h @@ -449,7 +449,7 @@ void vcc_insert_socket(struct sock *sk); static inline int atm_guess_pdu2truesize(int size) { - return (SKB_DATA_ALIGN(size) + sizeof(struct skb_shared_info)); + return SKB_DATA_ALIGN(size) + sizeof(struct skb_shared_info); } diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index fb6aa6070921..f16a01081e15 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -71,7 +71,7 @@ static inline int is_zero_ether_addr(const u8 *addr) */ static inline int is_multicast_ether_addr(const u8 *addr) { - return (0x01 & addr[0]); + return 0x01 & addr[0]; } /** @@ -82,7 +82,7 @@ static inline int is_multicast_ether_addr(const u8 *addr) */ static inline int is_local_ether_addr(const u8 *addr) { - return (0x02 & addr[0]); + return 0x02 & addr[0]; } /** diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index f7f1302138af..45dcda5bfda9 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1676,7 +1676,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index) */ static inline int netif_is_multiqueue(const struct net_device *dev) { - return (dev->num_tx_queues > 1); + return dev->num_tx_queues > 1; } extern void netif_set_real_num_tx_queues(struct net_device *dev, diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9e8085a89589..b2c41d19735c 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -601,7 +601,7 @@ static inline int skb_queue_empty(const struct sk_buff_head *list) static inline bool skb_queue_is_last(const struct sk_buff_head *list, const struct sk_buff *skb) { - return (skb->next == (struct sk_buff *) list); + return skb->next == (struct sk_buff *)list; } /** @@ -614,7 +614,7 @@ static inline bool skb_queue_is_last(const struct sk_buff_head *list, static inline bool skb_queue_is_first(const struct sk_buff_head *list, const struct sk_buff *skb) { - return (skb->prev == (struct sk_buff *) list); + return skb->prev == (struct sk_buff *)list; } /** @@ -2156,7 +2156,7 @@ static inline u16 skb_get_rx_queue(const struct sk_buff *skb) static inline bool skb_rx_queue_recorded(const struct sk_buff *skb) { - return (skb->queue_mapping != 0); + return skb->queue_mapping != 0; } extern u16 skb_tx_hash(const struct net_device *dev, diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 4568b938ca35..ebec8c9a929d 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -233,7 +233,7 @@ static inline void inquiry_cache_init(struct hci_dev *hdev) static inline int inquiry_cache_empty(struct hci_dev *hdev) { struct inquiry_cache *c = &hdev->inq_cache; - return (c->list == NULL); + return c->list == NULL; } static inline long inquiry_cache_age(struct hci_dev *hdev) diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h index 6c241444f902..c819c8bf9b68 100644 --- a/include/net/bluetooth/l2cap.h +++ b/include/net/bluetooth/l2cap.h @@ -414,7 +414,7 @@ static inline int l2cap_tx_window_full(struct sock *sk) if (sub < 0) sub += 64; - return (sub == pi->remote_tx_win); + return sub == pi->remote_tx_win; } #define __get_txseq(ctrl) ((ctrl) & L2CAP_CTRL_TXSEQ) >> 1 diff --git a/include/net/inet_ecn.h b/include/net/inet_ecn.h index 9b5d08f4f6e8..88bdd010d65d 100644 --- a/include/net/inet_ecn.h +++ b/include/net/inet_ecn.h @@ -27,7 +27,7 @@ static inline int INET_ECN_is_not_ect(__u8 dsfield) static inline int INET_ECN_is_capable(__u8 dsfield) { - return (dsfield & INET_ECN_ECT_0); + return dsfield & INET_ECN_ECT_0; } static inline __u8 INET_ECN_encapsulate(__u8 outer, __u8 inner) diff --git a/include/net/ip.h b/include/net/ip.h index 7691aca133db..dbee3fe260e1 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -238,9 +238,9 @@ int ip_decrease_ttl(struct iphdr *iph) static inline int ip_dont_fragment(struct sock *sk, struct dst_entry *dst) { - return (inet_sk(sk)->pmtudisc == IP_PMTUDISC_DO || + return inet_sk(sk)->pmtudisc == IP_PMTUDISC_DO || (inet_sk(sk)->pmtudisc == IP_PMTUDISC_WANT && - !(dst_metric_locked(dst, RTAX_MTU)))); + !(dst_metric_locked(dst, RTAX_MTU))); } extern void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst, int more); diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 1f8412410998..4a3cd2cd2f5e 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -262,7 +262,7 @@ static inline int ipv6_addr_scope(const struct in6_addr *addr) static inline int __ipv6_addr_src_scope(int type) { - return (type == IPV6_ADDR_ANY ? __IPV6_ADDR_SCOPE_INVALID : (type >> 16)); + return (type == IPV6_ADDR_ANY) ? __IPV6_ADDR_SCOPE_INVALID : (type >> 16); } static inline int ipv6_addr_src_scope(const struct in6_addr *addr) @@ -279,10 +279,10 @@ static inline int ipv6_masked_addr_cmp(const struct in6_addr *a1, const struct in6_addr *m, const struct in6_addr *a2) { - return (!!(((a1->s6_addr32[0] ^ a2->s6_addr32[0]) & m->s6_addr32[0]) | - ((a1->s6_addr32[1] ^ a2->s6_addr32[1]) & m->s6_addr32[1]) | - ((a1->s6_addr32[2] ^ a2->s6_addr32[2]) & m->s6_addr32[2]) | - ((a1->s6_addr32[3] ^ a2->s6_addr32[3]) & m->s6_addr32[3]))); + return !!(((a1->s6_addr32[0] ^ a2->s6_addr32[0]) & m->s6_addr32[0]) | + ((a1->s6_addr32[1] ^ a2->s6_addr32[1]) & m->s6_addr32[1]) | + ((a1->s6_addr32[2] ^ a2->s6_addr32[2]) & m->s6_addr32[2]) | + ((a1->s6_addr32[3] ^ a2->s6_addr32[3]) & m->s6_addr32[3])); } static inline void ipv6_addr_copy(struct in6_addr *a1, const struct in6_addr *a2) @@ -317,10 +317,10 @@ static inline void ipv6_addr_set(struct in6_addr *addr, static inline int ipv6_addr_equal(const struct in6_addr *a1, const struct in6_addr *a2) { - return (((a1->s6_addr32[0] ^ a2->s6_addr32[0]) | - (a1->s6_addr32[1] ^ a2->s6_addr32[1]) | - (a1->s6_addr32[2] ^ a2->s6_addr32[2]) | - (a1->s6_addr32[3] ^ a2->s6_addr32[3])) == 0); + return ((a1->s6_addr32[0] ^ a2->s6_addr32[0]) | + (a1->s6_addr32[1] ^ a2->s6_addr32[1]) | + (a1->s6_addr32[2] ^ a2->s6_addr32[2]) | + (a1->s6_addr32[3] ^ a2->s6_addr32[3])) == 0; } static inline int __ipv6_prefix_equal(const __be32 *a1, const __be32 *a2, @@ -373,20 +373,20 @@ int ip6_frag_match(struct inet_frag_queue *q, void *a); static inline int ipv6_addr_any(const struct in6_addr *a) { - return ((a->s6_addr32[0] | a->s6_addr32[1] | - a->s6_addr32[2] | a->s6_addr32[3] ) == 0); + return (a->s6_addr32[0] | a->s6_addr32[1] | + a->s6_addr32[2] | a->s6_addr32[3]) == 0; } static inline int ipv6_addr_loopback(const struct in6_addr *a) { - return ((a->s6_addr32[0] | a->s6_addr32[1] | - a->s6_addr32[2] | (a->s6_addr32[3] ^ htonl(1))) == 0); + return (a->s6_addr32[0] | a->s6_addr32[1] | + a->s6_addr32[2] | (a->s6_addr32[3] ^ htonl(1))) == 0; } static inline int ipv6_addr_v4mapped(const struct in6_addr *a) { - return ((a->s6_addr32[0] | a->s6_addr32[1] | - (a->s6_addr32[2] ^ htonl(0x0000ffff))) == 0); + return (a->s6_addr32[0] | a->s6_addr32[1] | + (a->s6_addr32[2] ^ htonl(0x0000ffff))) == 0; } /* @@ -395,8 +395,7 @@ static inline int ipv6_addr_v4mapped(const struct in6_addr *a) */ static inline int ipv6_addr_orchid(const struct in6_addr *a) { - return ((a->s6_addr32[0] & htonl(0xfffffff0)) - == htonl(0x20010010)); + return (a->s6_addr32[0] & htonl(0xfffffff0)) == htonl(0x20010010); } static inline void ipv6_addr_set_v4mapped(const __be32 addr, @@ -441,7 +440,7 @@ static inline int __ipv6_addr_diff(const void *token1, const void *token2, int a * if returned value is greater than prefix length. * --ANK (980803) */ - return (addrlen << 5); + return addrlen << 5; } static inline int ipv6_addr_diff(const struct in6_addr *a1, const struct in6_addr *a2) diff --git a/include/net/irda/irlap.h b/include/net/irda/irlap.h index 9d0c78ea92f5..17fcd964f9d9 100644 --- a/include/net/irda/irlap.h +++ b/include/net/irda/irlap.h @@ -282,7 +282,7 @@ static inline int irlap_is_primary(struct irlap_cb *self) default: ret = -1; } - return(ret); + return ret; } /* Clear a pending IrLAP disconnect. - Jean II */ diff --git a/include/net/irda/irlmp.h b/include/net/irda/irlmp.h index 3ffc1d0f93d6..fff11b7fe8a4 100644 --- a/include/net/irda/irlmp.h +++ b/include/net/irda/irlmp.h @@ -274,7 +274,7 @@ static inline int irlmp_lap_tx_queue_full(struct lsap_cb *self) if (self->lap->irlap == NULL) return 0; - return(IRLAP_GET_TX_QUEUE_LEN(self->lap->irlap) >= LAP_HIGH_THRESHOLD); + return IRLAP_GET_TX_QUEUE_LEN(self->lap->irlap) >= LAP_HIGH_THRESHOLD; } /* After doing a irlmp_dup(), this get one of the two socket back into diff --git a/include/net/irda/irttp.h b/include/net/irda/irttp.h index 11aee7a2972a..af4b87721d13 100644 --- a/include/net/irda/irttp.h +++ b/include/net/irda/irttp.h @@ -204,7 +204,7 @@ static inline int irttp_is_primary(struct tsap_cb *self) (self->lsap->lap == NULL) || (self->lsap->lap->irlap == NULL)) return -2; - return(irlap_is_primary(self->lsap->lap->irlap)); + return irlap_is_primary(self->lsap->lap->irlap); } #endif /* IRTTP_H */ diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 3c8728aaab4e..eda8808fdacd 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -601,7 +601,7 @@ static inline u32 qdisc_l2t(struct qdisc_rate_table* rtab, unsigned int pktlen) slot = 0; slot >>= rtab->rate.cell_log; if (slot > 255) - return (rtab->data[255]*(slot >> 8) + rtab->data[slot & 0xFF]); + return rtab->data[255]*(slot >> 8) + rtab->data[slot & 0xFF]; return rtab->data[slot]; } diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 2cb3980b1616..505845ddb0be 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -405,7 +405,7 @@ static inline void sctp_v6_del_protocol(void) { return; } /* Map an association to an assoc_id. */ static inline sctp_assoc_t sctp_assoc2id(const struct sctp_association *asoc) { - return (asoc?asoc->assoc_id:0); + return asoc ? asoc->assoc_id : 0; } /* Look up the association by its id. */ @@ -473,7 +473,7 @@ static inline void sctp_skb_set_owner_r(struct sk_buff *skb, struct sock *sk) /* Tests if the list has one and only one entry. */ static inline int sctp_list_single_entry(struct list_head *head) { - return ((head->next != head) && (head->next == head->prev)); + return (head->next != head) && (head->next == head->prev); } /* Generate a random jitter in the range of -50% ~ +50% of input RTO. */ @@ -631,13 +631,13 @@ static inline int sctp_sanity_check(void) /* This is the hash function for the SCTP port hash table. */ static inline int sctp_phashfn(__u16 lport) { - return (lport & (sctp_port_hashsize - 1)); + return lport & (sctp_port_hashsize - 1); } /* This is the hash function for the endpoint hash table. */ static inline int sctp_ep_hashfn(__u16 lport) { - return (lport & (sctp_ep_hashsize - 1)); + return lport & (sctp_ep_hashsize - 1); } /* This is the hash function for the association hash table. */ @@ -645,7 +645,7 @@ static inline int sctp_assoc_hashfn(__u16 lport, __u16 rport) { int h = (lport << 16) + rport; h ^= h>>8; - return (h & (sctp_assoc_hashsize - 1)); + return h & (sctp_assoc_hashsize - 1); } /* This is the hash function for the association hash table. This is @@ -656,7 +656,7 @@ static inline int sctp_vtag_hashfn(__u16 lport, __u16 rport, __u32 vtag) { int h = (lport << 16) + rport; h ^= vtag; - return (h & (sctp_assoc_hashsize-1)); + return h & (sctp_assoc_hashsize - 1); } #define sctp_for_each_hentry(epb, node, head) \ diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index 4088c89a9055..9352d12f02de 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -345,12 +345,12 @@ enum { static inline int TSN_lt(__u32 s, __u32 t) { - return (((s) - (t)) & TSN_SIGN_BIT); + return ((s) - (t)) & TSN_SIGN_BIT; } static inline int TSN_lte(__u32 s, __u32 t) { - return (((s) == (t)) || (((s) - (t)) & TSN_SIGN_BIT)); + return ((s) == (t)) || (((s) - (t)) & TSN_SIGN_BIT); } /* Compare two SSNs */ @@ -369,12 +369,12 @@ enum { static inline int SSN_lt(__u16 s, __u16 t) { - return (((s) - (t)) & SSN_SIGN_BIT); + return ((s) - (t)) & SSN_SIGN_BIT; } static inline int SSN_lte(__u16 s, __u16 t) { - return (((s) == (t)) || (((s) - (t)) & SSN_SIGN_BIT)); + return ((s) == (t)) || (((s) - (t)) & SSN_SIGN_BIT); } /* @@ -388,7 +388,7 @@ enum { static inline int ADDIP_SERIAL_gte(__u16 s, __u16 t) { - return (((s) == (t)) || (((t) - (s)) & ADDIP_SERIAL_SIGN_BIT)); + return ((s) == (t)) || (((t) - (s)) & ADDIP_SERIAL_SIGN_BIT); } /* Check VTAG of the packet matches the sender's own tag. */ diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index f9e7473613bd..69fef4fb79c0 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -847,7 +847,7 @@ void sctp_packet_free(struct sctp_packet *); static inline int sctp_packet_empty(struct sctp_packet *packet) { - return (packet->size == packet->overhead); + return packet->size == packet->overhead; } /* This represents a remote transport address. diff --git a/include/net/sctp/tsnmap.h b/include/net/sctp/tsnmap.h index 4aabc5a96cf6..e7728bc14ccf 100644 --- a/include/net/sctp/tsnmap.h +++ b/include/net/sctp/tsnmap.h @@ -157,7 +157,7 @@ __u16 sctp_tsnmap_pending(struct sctp_tsnmap *map); /* Is there a gap in the TSN map? */ static inline int sctp_tsnmap_has_gap(const struct sctp_tsnmap *map) { - return (map->cumulative_tsn_ack_point != map->max_tsn_seen); + return map->cumulative_tsn_ack_point != map->max_tsn_seen; } /* Mark a duplicate TSN. Note: limit the storage of duplicate TSN diff --git a/include/net/tipc/tipc_msg.h b/include/net/tipc/tipc_msg.h index 2e159a812f83..ffe50b4e7b93 100644 --- a/include/net/tipc/tipc_msg.h +++ b/include/net/tipc/tipc_msg.h @@ -107,7 +107,7 @@ static inline u32 msg_hdr_sz(struct tipc_msg *m) static inline int msg_short(struct tipc_msg *m) { - return (msg_hdr_sz(m) == 24); + return msg_hdr_sz(m) == 24; } static inline u32 msg_size(struct tipc_msg *m) @@ -117,7 +117,7 @@ static inline u32 msg_size(struct tipc_msg *m) static inline u32 msg_data_sz(struct tipc_msg *m) { - return (msg_size(m) - msg_hdr_sz(m)); + return msg_size(m) - msg_hdr_sz(m); } static inline unchar *msg_data(struct tipc_msg *m) @@ -132,17 +132,17 @@ static inline u32 msg_type(struct tipc_msg *m) static inline u32 msg_named(struct tipc_msg *m) { - return (msg_type(m) == TIPC_NAMED_MSG); + return msg_type(m) == TIPC_NAMED_MSG; } static inline u32 msg_mcast(struct tipc_msg *m) { - return (msg_type(m) == TIPC_MCAST_MSG); + return msg_type(m) == TIPC_MCAST_MSG; } static inline u32 msg_connected(struct tipc_msg *m) { - return (msg_type(m) == TIPC_CONN_MSG); + return msg_type(m) == TIPC_CONN_MSG; } static inline u32 msg_errcode(struct tipc_msg *m) diff --git a/net/802/fc.c b/net/802/fc.c index 34cf1ee014b8..1e49f2d4ea96 100644 --- a/net/802/fc.c +++ b/net/802/fc.c @@ -70,7 +70,7 @@ static int fc_header(struct sk_buff *skb, struct net_device *dev, if(daddr) { memcpy(fch->daddr,daddr,dev->addr_len); - return(hdr_len); + return hdr_len; } return -hdr_len; } diff --git a/net/802/fddi.c b/net/802/fddi.c index 3ef0ab0a543a..94b3ad08f39a 100644 --- a/net/802/fddi.c +++ b/net/802/fddi.c @@ -82,10 +82,10 @@ static int fddi_header(struct sk_buff *skb, struct net_device *dev, if (daddr != NULL) { memcpy(fddi->daddr, daddr, dev->addr_len); - return(hl); + return hl; } - return(-hl); + return -hl; } @@ -108,7 +108,7 @@ static int fddi_rebuild_header(struct sk_buff *skb) { printk("%s: Don't know how to resolve type %04X addresses.\n", skb->dev->name, ntohs(fddi->hdr.llc_snap.ethertype)); - return(0); + return 0; } } @@ -162,7 +162,7 @@ __be16 fddi_type_trans(struct sk_buff *skb, struct net_device *dev) /* Assume 802.2 SNAP frames, for now */ - return(type); + return type; } EXPORT_SYMBOL(fddi_type_trans); @@ -170,9 +170,9 @@ EXPORT_SYMBOL(fddi_type_trans); int fddi_change_mtu(struct net_device *dev, int new_mtu) { if ((new_mtu < FDDI_K_SNAP_HLEN) || (new_mtu > FDDI_K_SNAP_DLEN)) - return(-EINVAL); + return -EINVAL; dev->mtu = new_mtu; - return(0); + return 0; } EXPORT_SYMBOL(fddi_change_mtu); diff --git a/net/802/hippi.c b/net/802/hippi.c index cd3e8e929529..91aca8780fd0 100644 --- a/net/802/hippi.c +++ b/net/802/hippi.c @@ -152,7 +152,7 @@ int hippi_change_mtu(struct net_device *dev, int new_mtu) if ((new_mtu < 68) || (new_mtu > 65280)) return -EINVAL; dev->mtu = new_mtu; - return(0); + return 0; } EXPORT_SYMBOL(hippi_change_mtu); diff --git a/net/802/tr.c b/net/802/tr.c index 1c6e596074df..5e20cf8a074b 100644 --- a/net/802/tr.c +++ b/net/802/tr.c @@ -145,7 +145,7 @@ static int tr_header(struct sk_buff *skb, struct net_device *dev, { memcpy(trh->daddr,daddr,dev->addr_len); tr_source_route(skb, trh, dev); - return(hdr_len); + return hdr_len; } return -hdr_len; diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index 889f4ac4459a..0eb486d342dc 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -27,7 +27,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, else if (vlan_id) goto drop; - return (polling ? netif_receive_skb(skb) : netif_rx(skb)); + return polling ? netif_receive_skb(skb) : netif_rx(skb); drop: dev_kfree_skb_any(skb); diff --git a/net/9p/client.c b/net/9p/client.c index dc6f2f26d023..f34b9f510818 100644 --- a/net/9p/client.c +++ b/net/9p/client.c @@ -61,13 +61,13 @@ static const match_table_t tokens = { inline int p9_is_proto_dotl(struct p9_client *clnt) { - return (clnt->proto_version == p9_proto_2000L); + return clnt->proto_version == p9_proto_2000L; } EXPORT_SYMBOL(p9_is_proto_dotl); inline int p9_is_proto_dotu(struct p9_client *clnt) { - return (clnt->proto_version == p9_proto_2000u); + return clnt->proto_version == p9_proto_2000u; } EXPORT_SYMBOL(p9_is_proto_dotu); diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index 7dca91bb8c57..15ea84ba344e 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -179,13 +179,13 @@ static unsigned char rfcomm_crc_table[256] = { /* FCS on 2 bytes */ static inline u8 __fcs(u8 *data) { - return (0xff - __crc(data)); + return 0xff - __crc(data); } /* FCS on 3 bytes */ static inline u8 __fcs2(u8 *data) { - return (0xff - rfcomm_crc_table[__crc(data) ^ data[2]]); + return 0xff - rfcomm_crc_table[__crc(data) ^ data[2]]; } /* Check FCS */ diff --git a/net/core/flow.c b/net/core/flow.c index b143b86b1f2a..127c8a7ffd61 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -176,8 +176,8 @@ static u32 flow_hash_code(struct flow_cache *fc, { u32 *k = (u32 *) key; - return (jhash2(k, (sizeof(*key) / sizeof(u32)), fcp->hash_rnd) - & (flow_cache_hash_size(fc) - 1)); + return jhash2(k, (sizeof(*key) / sizeof(u32)), fcp->hash_rnd) + & (flow_cache_hash_size(fc) - 1); } typedef unsigned long flow_compare_t; diff --git a/net/core/neighbour.c b/net/core/neighbour.c index a4e0a7482c2b..96b1a749abb4 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -122,7 +122,7 @@ static void neigh_cleanup_and_release(struct neighbour *neigh) unsigned long neigh_rand_reach_time(unsigned long base) { - return (base ? (net_random() % base) + (base >> 1) : 0); + return base ? (net_random() % base) + (base >> 1) : 0; } EXPORT_SYMBOL(neigh_rand_reach_time); @@ -766,9 +766,9 @@ next_elt: static __inline__ int neigh_max_probes(struct neighbour *n) { struct neigh_parms *p = n->parms; - return (n->nud_state & NUD_PROBE ? + return (n->nud_state & NUD_PROBE) ? p->ucast_probes : - p->ucast_probes + p->app_probes + p->mcast_probes); + p->ucast_probes + p->app_probes + p->mcast_probes; } static void neigh_invalidate(struct neighbour *neigh) diff --git a/net/core/utils.c b/net/core/utils.c index ec6bb322f372..5fea0ab21902 100644 --- a/net/core/utils.c +++ b/net/core/utils.c @@ -75,7 +75,7 @@ __be32 in_aton(const char *str) str++; } } - return(htonl(l)); + return htonl(l); } EXPORT_SYMBOL(in_aton); diff --git a/net/dccp/ccids/lib/loss_interval.c b/net/dccp/ccids/lib/loss_interval.c index 8fc3cbf79071..497723c4d4bb 100644 --- a/net/dccp/ccids/lib/loss_interval.c +++ b/net/dccp/ccids/lib/loss_interval.c @@ -116,7 +116,7 @@ u8 tfrc_lh_update_i_mean(struct tfrc_loss_hist *lh, struct sk_buff *skb) cur->li_length = len; tfrc_lh_calc_i_mean(lh); - return (lh->i_mean < old_i_mean); + return lh->i_mean < old_i_mean; } /* Determine if `new_loss' does begin a new loss interval [RFC 4342, 10.2] */ diff --git a/net/econet/af_econet.c b/net/econet/af_econet.c index baa98fb83552..f8c1ae4b41f0 100644 --- a/net/econet/af_econet.c +++ b/net/econet/af_econet.c @@ -392,7 +392,7 @@ static int econet_sendmsg(struct kiocb *iocb, struct socket *sock, dev_queue_xmit(skb); dev_put(dev); mutex_unlock(&econet_mutex); - return(len); + return len; out_free: kfree_skb(skb); @@ -637,7 +637,7 @@ static int econet_create(struct net *net, struct socket *sock, int protocol, eo->num = protocol; econet_insert_socket(&econet_sklist, sk); - return(0); + return 0; out: return err; } diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c index 85e7b4551326..f00ef2f1d814 100644 --- a/net/ethernet/eth.c +++ b/net/ethernet/eth.c @@ -387,6 +387,6 @@ ssize_t sysfs_format_mac(char *buf, const unsigned char *addr, int len) l = _format_mac_addr(buf, PAGE_SIZE, addr, len); l += scnprintf(buf + l, PAGE_SIZE - l, "\n"); - return ((ssize_t) l); + return (ssize_t)l; } EXPORT_SYMBOL(sysfs_format_mac); diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index dcfe7e961c10..4083c186fd30 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -567,7 +567,7 @@ static inline int arp_fwd_proxy(struct in_device *in_dev, if (out_dev) omi = IN_DEV_MEDIUM_ID(out_dev); - return (omi != imi && omi != -1); + return omi != imi && omi != -1; } /* diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c index 721a8a37b45c..174be6caa5c8 100644 --- a/net/ipv4/datagram.c +++ b/net/ipv4/datagram.c @@ -73,6 +73,6 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) inet->inet_id = jiffies; sk_dst_set(sk, &rt->dst); - return(0); + return 0; } EXPORT_SYMBOL(ip4_datagram_connect); diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index e5fa2ddce320..ba8042665849 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -425,7 +425,7 @@ static int inet_diag_bc_run(const void *bc, int len, bc += op->no; } } - return (len == 0); + return len == 0; } static int valid_cc(const void *bc, int len, int cc) diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index f4dc879e258e..168440834ade 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -116,11 +116,11 @@ static int ip4_frag_match(struct inet_frag_queue *q, void *a) struct ip4_create_arg *arg = a; qp = container_of(q, struct ipq, q); - return (qp->id == arg->iph->id && + return qp->id == arg->iph->id && qp->saddr == arg->iph->saddr && qp->daddr == arg->iph->daddr && qp->protocol == arg->iph->protocol && - qp->user == arg->user); + qp->user == arg->user; } /* Memory Tracking Functions. */ diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 714b6a80361d..0967d02fefd8 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -659,7 +659,7 @@ drop: rcu_read_unlock(); drop_nolock: kfree_skb(skb); - return(0); + return 0; } static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index e8f4f9a57f12..8b642f152468 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -72,7 +72,7 @@ static inline int arp_devaddr_compare(const struct arpt_devaddr_info *ap, for (i = 0; i < len; i++) ret |= (hdr_addr[i] ^ ap->addr[i]) & ap->mask[i]; - return (ret != 0); + return ret != 0; } /* diff --git a/net/ipv4/route.c b/net/ipv4/route.c index e24d48dd99d3..ae1d4a41f1c6 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2791,7 +2791,7 @@ static int ipv4_dst_blackhole(struct net *net, struct rtable **rp, struct flowi dst_release(&(*rp)->dst); *rp = rt; - return (rt ? 0 : -ENOMEM); + return rt ? 0 : -ENOMEM; } int ip_route_output_flow(struct net *net, struct rtable **rp, struct flowi *flp, diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 1bc87a05c734..51966b3f9719 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2301,7 +2301,7 @@ static inline int tcp_dupack_heuristics(struct tcp_sock *tp) static inline int tcp_skb_timedout(struct sock *sk, struct sk_buff *skb) { - return (tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto); + return tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto; } static inline int tcp_head_timedout(struct sock *sk) @@ -3398,8 +3398,8 @@ static void tcp_ack_probe(struct sock *sk) static inline int tcp_ack_is_dubious(const struct sock *sk, const int flag) { - return (!(flag & FLAG_NOT_DUP) || (flag & FLAG_CA_ALERT) || - inet_csk(sk)->icsk_ca_state != TCP_CA_Open); + return !(flag & FLAG_NOT_DUP) || (flag & FLAG_CA_ALERT) || + inet_csk(sk)->icsk_ca_state != TCP_CA_Open; } static inline int tcp_may_raise_cwnd(const struct sock *sk, const int flag) @@ -3416,9 +3416,9 @@ static inline int tcp_may_update_window(const struct tcp_sock *tp, const u32 ack, const u32 ack_seq, const u32 nwin) { - return (after(ack, tp->snd_una) || + return after(ack, tp->snd_una) || after(ack_seq, tp->snd_wl1) || - (ack_seq == tp->snd_wl1 && nwin > tp->snd_wnd)); + (ack_seq == tp->snd_wl1 && nwin > tp->snd_wnd); } /* Update our send window. diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index f25b56cb85cb..43cf901d7659 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -55,7 +55,7 @@ static __inline__ int tcp_in_window(u32 seq, u32 end_seq, u32 s_win, u32 e_win) return 1; if (after(end_seq, s_win) && before(seq, e_win)) return 1; - return (seq == e_win && seq == end_seq); + return seq == e_win && seq == end_seq; } /* diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index ea09d2fd50c7..05b1ecf36763 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1370,9 +1370,9 @@ static inline int tcp_nagle_check(const struct tcp_sock *tp, const struct sk_buff *skb, unsigned mss_now, int nonagle) { - return (skb->len < mss_now && + return skb->len < mss_now && ((nonagle & TCP_NAGLE_CORK) || - (!nonagle && tp->packets_out && tcp_minshall_check(tp)))); + (!nonagle && tp->packets_out && tcp_minshall_check(tp))); } /* Return non-zero if the Nagle test allows this packet to be @@ -1443,10 +1443,10 @@ int tcp_may_send_now(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb = tcp_send_head(sk); - return (skb && + return skb && tcp_snd_test(sk, skb, tcp_current_mss(sk), (tcp_skb_is_last(sk, skb) ? - tp->nonagle : TCP_NAGLE_PUSH))); + tp->nonagle : TCP_NAGLE_PUSH)); } /* Trim TSO SKB to LEN bytes, put the remaining data into a new packet diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp_westwood.c index 20151d6a6241..a534dda5456e 100644 --- a/net/ipv4/tcp_westwood.c +++ b/net/ipv4/tcp_westwood.c @@ -80,7 +80,7 @@ static void tcp_westwood_init(struct sock *sk) */ static inline u32 westwood_do_filter(u32 a, u32 b) { - return (((7 * a) + b) >> 3); + return ((7 * a) + b) >> 3; } static void westwood_filter(struct westwood *w, u32 delta) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 5bc893e28008..89aa54394a08 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -243,7 +243,7 @@ static inline bool addrconf_qdisc_ok(const struct net_device *dev) /* Check if a route is valid prefix route */ static inline int addrconf_is_prefix_route(const struct rt6_info *rt) { - return ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0); + return (rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0; } static void addrconf_del_timer(struct inet6_ifaddr *ifp) diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c index f0e774cea386..921dcf6c271a 100644 --- a/net/ipv6/addrlabel.c +++ b/net/ipv6/addrlabel.c @@ -513,10 +513,9 @@ static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb) static inline int ip6addrlbl_msgsize(void) { - return (NLMSG_ALIGN(sizeof(struct ifaddrlblmsg)) + return NLMSG_ALIGN(sizeof(struct ifaddrlblmsg)) + nla_total_size(16) /* IFAL_ADDRESS */ - + nla_total_size(4) /* IFAL_LABEL */ - ); + + nla_total_size(4); /* IFAL_LABEL */ } static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh, diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 56b9bf2516f4..60220985bb80 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -467,7 +467,7 @@ int inet6_getname(struct socket *sock, struct sockaddr *uaddr, if (ipv6_addr_type(&sin->sin6_addr) & IPV6_ADDR_LINKLOCAL) sin->sin6_scope_id = sk->sk_bound_dev_if; *uaddr_len = sizeof(*sin); - return(0); + return 0; } EXPORT_SYMBOL(inet6_getname); @@ -488,7 +488,7 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) case SIOCADDRT: case SIOCDELRT: - return(ipv6_route_ioctl(net, cmd, (void __user *)arg)); + return ipv6_route_ioctl(net, cmd, (void __user *)arg); case SIOCSIFADDR: return addrconf_add_ifaddr(net, (void __user *) arg); @@ -502,7 +502,7 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) return sk->sk_prot->ioctl(sk, cmd, arg); } /*NOTREACHED*/ - return(0); + return 0; } EXPORT_SYMBOL(inet6_ioctl); diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c index e1caa5d526c2..14ed0a955b56 100644 --- a/net/ipv6/exthdrs_core.c +++ b/net/ipv6/exthdrs_core.c @@ -13,12 +13,12 @@ int ipv6_ext_hdr(u8 nexthdr) /* * find out if nexthdr is an extension header or a protocol */ - return ( (nexthdr == NEXTHDR_HOP) || + return (nexthdr == NEXTHDR_HOP) || (nexthdr == NEXTHDR_ROUTING) || (nexthdr == NEXTHDR_FRAGMENT) || (nexthdr == NEXTHDR_AUTH) || (nexthdr == NEXTHDR_NONE) || - (nexthdr == NEXTHDR_DEST) ); + (nexthdr == NEXTHDR_DEST); } /* diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 1838927a2243..efbbbce68f9e 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -870,8 +870,8 @@ static inline int ip6_rt_check(struct rt6key *rt_key, struct in6_addr *fl_addr, struct in6_addr *addr_cache) { - return ((rt_key->plen != 128 || !ipv6_addr_equal(fl_addr, &rt_key->addr)) && - (addr_cache == NULL || !ipv6_addr_equal(fl_addr, addr_cache))); + return (rt_key->plen != 128 || !ipv6_addr_equal(fl_addr, &rt_key->addr)) && + (addr_cache == NULL || !ipv6_addr_equal(fl_addr, addr_cache)); } static struct dst_entry *ip6_sk_dst_check(struct sock *sk, diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 69a0051cea67..b3dd844cd34f 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -228,12 +228,12 @@ static struct nd_opt_hdr *ndisc_next_option(struct nd_opt_hdr *cur, do { cur = ((void *)cur) + (cur->nd_opt_len << 3); } while(cur < end && cur->nd_opt_type != type); - return (cur <= end && cur->nd_opt_type == type ? cur : NULL); + return cur <= end && cur->nd_opt_type == type ? cur : NULL; } static inline int ndisc_is_useropt(struct nd_opt_hdr *opt) { - return (opt->nd_opt_type == ND_OPT_RDNSS); + return opt->nd_opt_type == ND_OPT_RDNSS; } static struct nd_opt_hdr *ndisc_next_useropt(struct nd_opt_hdr *cur, @@ -244,7 +244,7 @@ static struct nd_opt_hdr *ndisc_next_useropt(struct nd_opt_hdr *cur, do { cur = ((void *)cur) + (cur->nd_opt_len << 3); } while(cur < end && !ndisc_is_useropt(cur)); - return (cur <= end && ndisc_is_useropt(cur) ? cur : NULL); + return cur <= end && ndisc_is_useropt(cur) ? cur : NULL; } static struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, @@ -319,7 +319,7 @@ static inline u8 *ndisc_opt_addr_data(struct nd_opt_hdr *p, int prepad = ndisc_addr_option_pad(dev->type); if (lladdrlen != NDISC_OPT_SPACE(dev->addr_len + prepad)) return NULL; - return (lladdr + prepad); + return lladdr + prepad; } int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int dir) diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 8e754be92c24..6b331e9b5706 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -82,13 +82,13 @@ EXPORT_SYMBOL_GPL(ip6t_alloc_initial_table); int ip6t_ext_hdr(u8 nexthdr) { - return ( (nexthdr == IPPROTO_HOPOPTS) || - (nexthdr == IPPROTO_ROUTING) || - (nexthdr == IPPROTO_FRAGMENT) || - (nexthdr == IPPROTO_ESP) || - (nexthdr == IPPROTO_AH) || - (nexthdr == IPPROTO_NONE) || - (nexthdr == IPPROTO_DSTOPTS) ); + return (nexthdr == IPPROTO_HOPOPTS) || + (nexthdr == IPPROTO_ROUTING) || + (nexthdr == IPPROTO_FRAGMENT) || + (nexthdr == IPPROTO_ESP) || + (nexthdr == IPPROTO_AH) || + (nexthdr == IPPROTO_NONE) || + (nexthdr == IPPROTO_DSTOPTS); } /* Returns whether matches rule or not. */ diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index e677937a07fc..45e6efb7f171 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -764,7 +764,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, return -EINVAL; if (sin6->sin6_family && sin6->sin6_family != AF_INET6) - return(-EAFNOSUPPORT); + return -EAFNOSUPPORT; /* port is the proto value [0..255] carried in nexthdr */ proto = ntohs(sin6->sin6_port); @@ -772,10 +772,10 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, if (!proto) proto = inet->inet_num; else if (proto != inet->inet_num) - return(-EINVAL); + return -EINVAL; if (proto > 255) - return(-EINVAL); + return -EINVAL; daddr = &sin6->sin6_addr; if (np->sndflow) { @@ -985,7 +985,7 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname, /* You may get strange result with a positive odd offset; RFC2292bis agrees with me. */ if (val > 0 && (val&1)) - return(-EINVAL); + return -EINVAL; if (val < 0) { rp->checksum = 0; } else { @@ -997,7 +997,7 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname, break; default: - return(-ENOPROTOOPT); + return -ENOPROTOOPT; } } @@ -1190,7 +1190,7 @@ static int rawv6_init_sk(struct sock *sk) default: break; } - return(0); + return 0; } struct proto rawv6_prot = { diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d126365ac046..25b0beda4331 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -217,14 +217,14 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, static __inline__ int rt6_check_expired(const struct rt6_info *rt) { - return (rt->rt6i_flags & RTF_EXPIRES && - time_after(jiffies, rt->rt6i_expires)); + return (rt->rt6i_flags & RTF_EXPIRES) && + time_after(jiffies, rt->rt6i_expires); } static inline int rt6_need_strict(struct in6_addr *daddr) { - return (ipv6_addr_type(daddr) & - (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK)); + return ipv6_addr_type(daddr) & + (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK); } /* @@ -440,7 +440,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict) __func__, match); net = dev_net(rt0->rt6i_dev); - return (match ? match : net->ipv6.ip6_null_entry); + return match ? match : net->ipv6.ip6_null_entry; } #ifdef CONFIG_IPV6_ROUTE_INFO @@ -859,7 +859,7 @@ int ip6_dst_blackhole(struct sock *sk, struct dst_entry **dstp, struct flowi *fl dst_release(*dstp); *dstp = new; - return (new ? 0 : -ENOMEM); + return new ? 0 : -ENOMEM; } EXPORT_SYMBOL_GPL(ip6_dst_blackhole); @@ -1070,7 +1070,7 @@ static int ip6_dst_gc(struct dst_ops *ops) net->ipv6.ip6_rt_gc_expire = rt_gc_timeout>>1; out: net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>rt_elasticity; - return (atomic_read(&ops->entries) > rt_max_size); + return atomic_read(&ops->entries) > rt_max_size; } /* Clean host part of a prefix. Not necessary in radix tree, diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index fe6d40418c0b..8d93f6d81979 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -139,7 +139,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, return -EINVAL; if (usin->sin6_family != AF_INET6) - return(-EAFNOSUPPORT); + return -EAFNOSUPPORT; memset(&fl, 0, sizeof(fl)); diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 6baeabbbca82..39676eac3a37 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -199,7 +199,7 @@ static inline int xfrm6_garbage_collect(struct dst_ops *ops) struct net *net = container_of(ops, struct net, xfrm.xfrm6_dst_ops); xfrm6_policy_afinfo.garbage_collect(net); - return (atomic_read(&ops->entries) > ops->gc_thresh * 2); + return atomic_read(&ops->entries) > ops->gc_thresh * 2; } static void xfrm6_update_pmtu(struct dst_entry *dst, u32 mtu) diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c index fd55b5135de5..bf3635129b17 100644 --- a/net/irda/af_irda.c +++ b/net/irda/af_irda.c @@ -573,9 +573,9 @@ static int irda_find_lsap_sel(struct irda_sock *self, char *name) /* Requested object/attribute doesn't exist */ if((self->errno == IAS_CLASS_UNKNOWN) || (self->errno == IAS_ATTRIB_UNKNOWN)) - return (-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; else - return (-EHOSTUNREACH); + return -EHOSTUNREACH; } /* Get the remote TSAP selector */ @@ -663,7 +663,7 @@ static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name) __func__, name); self->daddr = DEV_ADDR_ANY; kfree(discoveries); - return(-ENOTUNIQ); + return -ENOTUNIQ; } /* First time we found that one, save it ! */ daddr = self->daddr; @@ -677,7 +677,7 @@ static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name) IRDA_DEBUG(0, "%s(), unexpected IAS query failure\n", __func__); self->daddr = DEV_ADDR_ANY; kfree(discoveries); - return(-EHOSTUNREACH); + return -EHOSTUNREACH; break; } } @@ -689,7 +689,7 @@ static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name) IRDA_DEBUG(1, "%s(), cannot discover service ''%s'' in any device !!!\n", __func__, name); self->daddr = DEV_ADDR_ANY; - return(-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; } /* Revert back to discovered device & service */ @@ -2465,9 +2465,9 @@ bed: /* Requested object/attribute doesn't exist */ if((self->errno == IAS_CLASS_UNKNOWN) || (self->errno == IAS_ATTRIB_UNKNOWN)) - return (-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; else - return (-EHOSTUNREACH); + return -EHOSTUNREACH; } /* Translate from internal to user structure */ diff --git a/net/irda/discovery.c b/net/irda/discovery.c index c1c8ae939126..36c3f037f172 100644 --- a/net/irda/discovery.c +++ b/net/irda/discovery.c @@ -315,7 +315,7 @@ struct irda_device_info *irlmp_copy_discoveries(hashbin_t *log, int *pn, /* Get the actual number of device in the buffer and return */ *pn = i; - return(buffer); + return buffer; } #ifdef CONFIG_PROC_FS diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c index faa82ca2dfdc..a39cca8331df 100644 --- a/net/irda/ircomm/ircomm_tty.c +++ b/net/irda/ircomm/ircomm_tty.c @@ -449,8 +449,8 @@ static int ircomm_tty_open(struct tty_struct *tty, struct file *filp) } #ifdef SERIAL_DO_RESTART - return ((self->flags & ASYNC_HUP_NOTIFY) ? - -EAGAIN : -ERESTARTSYS); + return (self->flags & ASYNC_HUP_NOTIFY) ? + -EAGAIN : -ERESTARTSYS; #else return -EAGAIN; #endif diff --git a/net/irda/irlmp.c b/net/irda/irlmp.c index 0e7d8bde145d..6115a44c0a24 100644 --- a/net/irda/irlmp.c +++ b/net/irda/irlmp.c @@ -939,7 +939,7 @@ struct irda_device_info *irlmp_get_discoveries(int *pn, __u16 mask, int nslots) } /* Return current cached discovery log */ - return(irlmp_copy_discoveries(irlmp->cachelog, pn, mask, TRUE)); + return irlmp_copy_discoveries(irlmp->cachelog, pn, mask, TRUE); } EXPORT_SYMBOL(irlmp_get_discoveries); diff --git a/net/irda/irlmp_frame.c b/net/irda/irlmp_frame.c index 3750884094da..062e63b1c5c4 100644 --- a/net/irda/irlmp_frame.c +++ b/net/irda/irlmp_frame.c @@ -448,7 +448,7 @@ static struct lsap_cb *irlmp_find_lsap(struct lap_cb *self, __u8 dlsap_sel, (self->cache.slsap_sel == slsap_sel) && (self->cache.dlsap_sel == dlsap_sel)) { - return (self->cache.lsap); + return self->cache.lsap; } #endif diff --git a/net/irda/irnet/irnet_irda.c b/net/irda/irnet/irnet_irda.c index e98e40d76f4f..7f17a8020e8a 100644 --- a/net/irda/irnet/irnet_irda.c +++ b/net/irda/irnet/irnet_irda.c @@ -238,7 +238,7 @@ irnet_ias_to_tsap(irnet_socket * self, DEXIT(IRDA_SR_TRACE, "\n"); /* Return the TSAP */ - return(dtsap_sel); + return dtsap_sel; } /*------------------------------------------------------------------*/ @@ -301,7 +301,7 @@ irnet_connect_tsap(irnet_socket * self) { clear_bit(0, &self->ttp_connect); DERROR(IRDA_SR_ERROR, "connect aborted!\n"); - return(err); + return err; } /* Connect to remote device */ @@ -312,7 +312,7 @@ irnet_connect_tsap(irnet_socket * self) { clear_bit(0, &self->ttp_connect); DERROR(IRDA_SR_ERROR, "connect aborted!\n"); - return(err); + return err; } /* The above call is non-blocking. @@ -321,7 +321,7 @@ irnet_connect_tsap(irnet_socket * self) * See you there ;-) */ DEXIT(IRDA_SR_TRACE, "\n"); - return(err); + return err; } /*------------------------------------------------------------------*/ @@ -362,10 +362,10 @@ irnet_discover_next_daddr(irnet_socket * self) /* The above request is non-blocking. * After a while, IrDA will call us back in irnet_discovervalue_confirm() * We will then call irnet_ias_to_tsap() and come back here again... */ - return(0); + return 0; } else - return(1); + return 1; } /*------------------------------------------------------------------*/ @@ -436,7 +436,7 @@ irnet_discover_daddr_and_lsap_sel(irnet_socket * self) /* Follow me in irnet_discovervalue_confirm() */ DEXIT(IRDA_SR_TRACE, "\n"); - return(0); + return 0; } /*------------------------------------------------------------------*/ @@ -485,7 +485,7 @@ irnet_dname_to_daddr(irnet_socket * self) /* No luck ! */ DEBUG(IRDA_SR_INFO, "cannot discover device ``%s'' !!!\n", self->rname); kfree(discoveries); - return(-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; } @@ -527,7 +527,7 @@ irda_irnet_create(irnet_socket * self) INIT_WORK(&self->disconnect_work, irnet_ppp_disconnect); DEXIT(IRDA_SOCK_TRACE, "\n"); - return(0); + return 0; } /*------------------------------------------------------------------*/ @@ -601,7 +601,7 @@ irda_irnet_connect(irnet_socket * self) * We will finish the connection procedure in irnet_connect_tsap(). */ DEXIT(IRDA_SOCK_TRACE, "\n"); - return(0); + return 0; } /*------------------------------------------------------------------*/ @@ -733,7 +733,7 @@ irnet_daddr_to_dname(irnet_socket * self) /* No luck ! */ DEXIT(IRDA_SERV_INFO, ": cannot discover device 0x%08x !!!\n", self->daddr); kfree(discoveries); - return(-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; } /*------------------------------------------------------------------*/ diff --git a/net/irda/irnet/irnet_ppp.c b/net/irda/irnet/irnet_ppp.c index dfe7b38dd4af..69f1fa64994e 100644 --- a/net/irda/irnet/irnet_ppp.c +++ b/net/irda/irnet/irnet_ppp.c @@ -166,7 +166,7 @@ irnet_ctrl_write(irnet_socket * ap, } /* Success : we have parsed all commands successfully */ - return(count); + return count; } #ifdef INITIAL_DISCOVERY @@ -300,7 +300,7 @@ irnet_ctrl_read(irnet_socket * ap, } DEXIT(CTRL_TRACE, "\n"); - return(strlen(event)); + return strlen(event); } #endif /* INITIAL_DISCOVERY */ @@ -409,7 +409,7 @@ irnet_ctrl_read(irnet_socket * ap, } DEXIT(CTRL_TRACE, "\n"); - return(strlen(event)); + return strlen(event); } /*------------------------------------------------------------------*/ @@ -623,7 +623,7 @@ dev_irnet_poll(struct file * file, mask |= irnet_ctrl_poll(ap, file, wait); DEXIT(FS_TRACE, " - mask=0x%X\n", mask); - return(mask); + return mask; } /*------------------------------------------------------------------*/ diff --git a/net/key/af_key.c b/net/key/af_key.c index 43040e97c474..d87c22df6f1e 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -565,12 +565,12 @@ pfkey_proto2satype(uint16_t proto) static uint8_t pfkey_proto_to_xfrm(uint8_t proto) { - return (proto == IPSEC_PROTO_ANY ? 0 : proto); + return proto == IPSEC_PROTO_ANY ? 0 : proto; } static uint8_t pfkey_proto_from_xfrm(uint8_t proto) { - return (proto ? proto : IPSEC_PROTO_ANY); + return proto ? proto : IPSEC_PROTO_ANY; } static inline int pfkey_sockaddr_len(sa_family_t family) diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c index 4f772de2f213..b0cc385bf989 100644 --- a/net/mac80211/rate.c +++ b/net/mac80211/rate.c @@ -207,7 +207,7 @@ static bool rc_no_data_or_no_ack(struct ieee80211_tx_rate_control *txrc) fc = hdr->frame_control; - return ((info->flags & IEEE80211_TX_CTL_NO_ACK) || !ieee80211_is_data(fc)); + return (info->flags & IEEE80211_TX_CTL_NO_ACK) || !ieee80211_is_data(fc); } static void rc_send_low_broadcast(s8 *idx, u32 basic_rates, u8 max_rate_idx) diff --git a/net/rfkill/input.c b/net/rfkill/input.c index 3713d7ecab96..1bca6d49ec96 100644 --- a/net/rfkill/input.c +++ b/net/rfkill/input.c @@ -142,7 +142,7 @@ static unsigned long rfkill_last_scheduled; static unsigned long rfkill_ratelimit(const unsigned long last) { const unsigned long delay = msecs_to_jiffies(RFKILL_OPS_DELAY); - return (time_after(jiffies, last + delay)) ? 0 : delay; + return time_after(jiffies, last + delay) ? 0 : delay; } static void rfkill_schedule_ratelimited(void) diff --git a/net/rose/rose_link.c b/net/rose/rose_link.c index a750a28e0221..fa5f5641a2c2 100644 --- a/net/rose/rose_link.c +++ b/net/rose/rose_link.c @@ -114,7 +114,7 @@ static int rose_send_frame(struct sk_buff *skb, struct rose_neigh *neigh) if (ax25s) ax25_cb_put(ax25s); - return (neigh->ax25 != NULL); + return neigh->ax25 != NULL; } /* @@ -137,7 +137,7 @@ static int rose_link_up(struct rose_neigh *neigh) if (ax25s) ax25_cb_put(ax25s); - return (neigh->ax25 != NULL); + return neigh->ax25 != NULL; } /* diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index f774e657641a..1ef29c74d85e 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -799,7 +799,7 @@ static void sctp_inet_skb_msgname(struct sk_buff *skb, char *msgname, int *len) static int sctp_inet_af_supported(sa_family_t family, struct sctp_sock *sp) { /* PF_INET only supports AF_INET addresses. */ - return (AF_INET == family); + return AF_INET == family; } /* Address matching with wildcards allowed. */ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 6a691d84aef4..535659fdbaa1 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3884,7 +3884,7 @@ static int sctp_getsockopt_sctp_status(struct sock *sk, int len, } out: - return (retval); + return retval; } @@ -3940,7 +3940,7 @@ static int sctp_getsockopt_peer_addr_info(struct sock *sk, int len, } out: - return (retval); + return retval; } /* 7.1.12 Enable/Disable message fragmentation (SCTP_DISABLE_FRAGMENTS) @@ -5594,7 +5594,7 @@ static int sctp_get_port(struct sock *sk, unsigned short snum) /* Note: sk->sk_num gets filled in if ephemeral port request. */ ret = sctp_get_port_local(sk, &addr); - return (ret ? 1 : 0); + return ret ? 1 : 0; } /* diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index dcfc66bab2bb..597c493392ad 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -1049,7 +1049,7 @@ gss_match(struct auth_cred *acred, struct rpc_cred *rc, int flags) out: if (acred->machine_cred != gss_cred->gc_machine_cred) return 0; - return (rc->cr_uid == acred->uid); + return rc->cr_uid == acred->uid; } /* diff --git a/net/sunrpc/auth_gss/gss_generic_token.c b/net/sunrpc/auth_gss/gss_generic_token.c index 310b78e99456..c586e92bcf76 100644 --- a/net/sunrpc/auth_gss/gss_generic_token.c +++ b/net/sunrpc/auth_gss/gss_generic_token.c @@ -76,19 +76,19 @@ static int der_length_size( int length) { if (length < (1<<7)) - return(1); + return 1; else if (length < (1<<8)) - return(2); + return 2; #if (SIZEOF_INT == 2) else - return(3); + return 3; #else else if (length < (1<<16)) - return(3); + return 3; else if (length < (1<<24)) - return(4); + return 4; else - return(5); + return 5; #endif } @@ -121,14 +121,14 @@ der_read_length(unsigned char **buf, int *bufsize) int ret; if (*bufsize < 1) - return(-1); + return -1; sf = *(*buf)++; (*bufsize)--; if (sf & 0x80) { if ((sf &= 0x7f) > ((*bufsize)-1)) - return(-1); + return -1; if (sf > SIZEOF_INT) - return (-1); + return -1; ret = 0; for (; sf; sf--) { ret = (ret<<8) + (*(*buf)++); @@ -138,7 +138,7 @@ der_read_length(unsigned char **buf, int *bufsize) ret = sf; } - return(ret); + return ret; } /* returns the length of a token, given the mech oid and the body size */ @@ -148,7 +148,7 @@ g_token_size(struct xdr_netobj *mech, unsigned int body_size) { /* set body_size to sequence contents size */ body_size += 2 + (int) mech->len; /* NEED overflow check */ - return(1 + der_length_size(body_size) + body_size); + return 1 + der_length_size(body_size) + body_size; } EXPORT_SYMBOL_GPL(g_token_size); @@ -186,27 +186,27 @@ g_verify_token_header(struct xdr_netobj *mech, int *body_size, int ret = 0; if ((toksize-=1) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if (*buf++ != 0x60) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if ((seqsize = der_read_length(&buf, &toksize)) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if (seqsize != toksize) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if ((toksize-=1) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if (*buf++ != 0x06) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if ((toksize-=1) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; toid.len = *buf++; if ((toksize-=toid.len) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; toid.data = buf; buf+=toid.len; @@ -217,17 +217,17 @@ g_verify_token_header(struct xdr_netobj *mech, int *body_size, to return G_BAD_TOK_HEADER if the token header is in fact bad */ if ((toksize-=2) < 0) - return(G_BAD_TOK_HEADER); + return G_BAD_TOK_HEADER; if (ret) - return(ret); + return ret; if (!ret) { *buf_in = buf; *body_size = toksize; } - return(ret); + return ret; } EXPORT_SYMBOL_GPL(g_verify_token_header); diff --git a/net/sunrpc/auth_gss/gss_krb5_seqnum.c b/net/sunrpc/auth_gss/gss_krb5_seqnum.c index 415c013ba382..62ac90c62cb1 100644 --- a/net/sunrpc/auth_gss/gss_krb5_seqnum.c +++ b/net/sunrpc/auth_gss/gss_krb5_seqnum.c @@ -162,5 +162,5 @@ krb5_get_seq_num(struct krb5_ctx *kctx, *seqnum = ((plain[0]) | (plain[1] << 8) | (plain[2] << 16) | (plain[3] << 24)); - return (0); + return 0; } diff --git a/net/sunrpc/auth_gss/gss_mech_switch.c b/net/sunrpc/auth_gss/gss_mech_switch.c index 2689de39dc78..8b4061049d76 100644 --- a/net/sunrpc/auth_gss/gss_mech_switch.c +++ b/net/sunrpc/auth_gss/gss_mech_switch.c @@ -331,7 +331,7 @@ gss_delete_sec_context(struct gss_ctx **context_handle) *context_handle); if (!*context_handle) - return(GSS_S_NO_CONTEXT); + return GSS_S_NO_CONTEXT; if ((*context_handle)->internal_ctx_id) (*context_handle)->mech_type->gm_ops ->gss_delete_sec_context((*context_handle) diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index cace6049e4a5..aa5dbda6608c 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -376,7 +376,7 @@ int rpc_queue_empty(struct rpc_wait_queue *queue) spin_lock_bh(&queue->lock); res = queue->qlen; spin_unlock_bh(&queue->lock); - return (res == 0); + return res == 0; } EXPORT_SYMBOL_GPL(rpc_queue_empty); diff --git a/net/tipc/addr.c b/net/tipc/addr.c index c048543ffbeb..2ddc351b3be9 100644 --- a/net/tipc/addr.c +++ b/net/tipc/addr.c @@ -89,7 +89,7 @@ int tipc_addr_domain_valid(u32 addr) int tipc_addr_node_valid(u32 addr) { - return (tipc_addr_domain_valid(addr) && tipc_node(addr)); + return tipc_addr_domain_valid(addr) && tipc_node(addr); } int tipc_in_scope(u32 domain, u32 addr) diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index b11248c2d788..ecfaac10d0b4 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -184,7 +184,7 @@ static void bclink_set_gap(struct tipc_node *n_ptr) static int bclink_ack_allowed(u32 n) { - return((n % TIPC_MIN_LINK_WIN) == tipc_own_tag); + return (n % TIPC_MIN_LINK_WIN) == tipc_own_tag; } diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 52ae17b2583e..9c10c6b7c12b 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -63,7 +63,7 @@ static int media_name_valid(const char *name) len = strlen(name); if ((len + 1) > TIPC_MAX_MEDIA_NAME) return 0; - return (strspn(name, tipc_alphabet) == len); + return strspn(name, tipc_alphabet) == len; } /** diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c index 1885a7edb0c8..6569d45bfb9a 100644 --- a/net/tipc/dbg.c +++ b/net/tipc/dbg.c @@ -134,7 +134,7 @@ void tipc_printbuf_reset(struct print_buf *pb) int tipc_printbuf_empty(struct print_buf *pb) { - return (!pb->buf || (pb->crs == pb->buf)); + return !pb->buf || (pb->crs == pb->buf); } /** @@ -169,7 +169,7 @@ int tipc_printbuf_validate(struct print_buf *pb) tipc_printf(pb, err); } } - return (pb->crs - pb->buf + 1); + return pb->crs - pb->buf + 1; } /** diff --git a/net/tipc/link.c b/net/tipc/link.c index a6a3102bb4d6..b8cf1e9d0b86 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -239,13 +239,13 @@ int tipc_link_is_up(struct link *l_ptr) { if (!l_ptr) return 0; - return (link_working_working(l_ptr) || link_working_unknown(l_ptr)); + return link_working_working(l_ptr) || link_working_unknown(l_ptr); } int tipc_link_is_active(struct link *l_ptr) { - return ((l_ptr->owner->active_links[0] == l_ptr) || - (l_ptr->owner->active_links[1] == l_ptr)); + return (l_ptr->owner->active_links[0] == l_ptr) || + (l_ptr->owner->active_links[1] == l_ptr); } /** diff --git a/net/tipc/link.h b/net/tipc/link.h index 2e5385c47d30..26151d30589d 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -279,12 +279,12 @@ static inline int between(u32 lower, u32 upper, u32 n) static inline int less_eq(u32 left, u32 right) { - return (mod(right - left) < 32768u); + return mod(right - left) < 32768u; } static inline int less(u32 left, u32 right) { - return (less_eq(left, right) && (mod(right) != mod(left))); + return less_eq(left, right) && (mod(right) != mod(left)); } static inline u32 lesser(u32 left, u32 right) @@ -299,32 +299,32 @@ static inline u32 lesser(u32 left, u32 right) static inline int link_working_working(struct link *l_ptr) { - return (l_ptr->state == WORKING_WORKING); + return l_ptr->state == WORKING_WORKING; } static inline int link_working_unknown(struct link *l_ptr) { - return (l_ptr->state == WORKING_UNKNOWN); + return l_ptr->state == WORKING_UNKNOWN; } static inline int link_reset_unknown(struct link *l_ptr) { - return (l_ptr->state == RESET_UNKNOWN); + return l_ptr->state == RESET_UNKNOWN; } static inline int link_reset_reset(struct link *l_ptr) { - return (l_ptr->state == RESET_RESET); + return l_ptr->state == RESET_RESET; } static inline int link_blocked(struct link *l_ptr) { - return (l_ptr->exp_msg_count || l_ptr->blocked); + return l_ptr->exp_msg_count || l_ptr->blocked; } static inline int link_congested(struct link *l_ptr) { - return (l_ptr->out_queue_size >= l_ptr->queue_limit[0]); + return l_ptr->out_queue_size >= l_ptr->queue_limit[0]; } #endif diff --git a/net/tipc/msg.h b/net/tipc/msg.h index 995d2da35b01..031aad18efce 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -104,7 +104,7 @@ static inline u32 msg_user(struct tipc_msg *m) static inline u32 msg_isdata(struct tipc_msg *m) { - return (msg_user(m) <= TIPC_CRITICAL_IMPORTANCE); + return msg_user(m) <= TIPC_CRITICAL_IMPORTANCE; } static inline void msg_set_user(struct tipc_msg *m, u32 n) @@ -289,7 +289,7 @@ static inline void msg_set_destnode(struct tipc_msg *m, u32 a) static inline int msg_is_dest(struct tipc_msg *m, u32 d) { - return(msg_short(m) || (msg_destnode(m) == d)); + return msg_short(m) || (msg_destnode(m) == d); } static inline u32 msg_routed(struct tipc_msg *m) @@ -632,7 +632,7 @@ static inline void msg_set_bcast_tag(struct tipc_msg *m, u32 n) static inline u32 msg_max_pkt(struct tipc_msg *m) { - return (msg_bits(m, 9, 16, 0xffff) * 4); + return msg_bits(m, 9, 16, 0xffff) * 4; } static inline void msg_set_max_pkt(struct tipc_msg *m, u32 n) diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c index c13c2c7c4b57..9ca4b0689237 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -116,7 +116,7 @@ DEFINE_RWLOCK(tipc_nametbl_lock); static int hash(int x) { - return(x & (tipc_nametbl_size - 1)); + return x & (tipc_nametbl_size - 1); } /** diff --git a/net/tipc/node.c b/net/tipc/node.c index b702c7bf580f..7c49cd056df7 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -242,17 +242,17 @@ int tipc_node_has_active_links(struct tipc_node *n_ptr) int tipc_node_has_redundant_links(struct tipc_node *n_ptr) { - return (n_ptr->working_links > 1); + return n_ptr->working_links > 1; } static int tipc_node_has_active_routes(struct tipc_node *n_ptr) { - return (n_ptr && (n_ptr->last_router >= 0)); + return n_ptr && (n_ptr->last_router >= 0); } int tipc_node_is_up(struct tipc_node *n_ptr) { - return (tipc_node_has_active_links(n_ptr) || tipc_node_has_active_routes(n_ptr)); + return tipc_node_has_active_links(n_ptr) || tipc_node_has_active_routes(n_ptr); } struct tipc_node *tipc_node_attach_link(struct link *l_ptr) diff --git a/net/tipc/port.h b/net/tipc/port.h index 8d1652aab298..e74bd9563739 100644 --- a/net/tipc/port.h +++ b/net/tipc/port.h @@ -157,7 +157,7 @@ static inline u32 tipc_peer_node(struct port *p_ptr) static inline int tipc_port_congested(struct port *p_ptr) { - return((p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2)); + return (p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2); } /** diff --git a/net/tipc/socket.c b/net/tipc/socket.c index f7ac94de24fe..33217fc3d697 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1195,7 +1195,7 @@ static int rx_queue_full(struct tipc_msg *msg, u32 queue_size, u32 base) if (msg_connected(msg)) threshold *= 4; - return (queue_size >= threshold); + return queue_size >= threshold; } /** diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index ab6eab4c45e2..1a5b9a6bd128 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -604,6 +604,6 @@ int tipc_ispublished(struct tipc_name const *name) { u32 domain = 0; - return(tipc_nametbl_translate(name->type, name->instance,&domain) != 0); + return tipc_nametbl_translate(name->type, name->instance, &domain) != 0; } diff --git a/net/wireless/core.h b/net/wireless/core.h index 37580e090a3d..5d89310b3587 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -86,7 +86,7 @@ struct cfg80211_registered_device *wiphy_to_dev(struct wiphy *wiphy) static inline bool wiphy_idx_valid(int wiphy_idx) { - return (wiphy_idx >= 0); + return wiphy_idx >= 0; } -- cgit v1.2.3 From c5256c51232d8312755e8de2b514c426b19b101a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 23 Sep 2010 00:46:11 +0000 Subject: net: propagate NETIF_F_HIGHDMA to vlans Automatically allows vlans to get NETIF_F_HIGHDMA if underlying device supports it. On 32bit arches (and more precisely if CONFIG_HIGHMEM is enabled), it can help to reduce cost of illegal_highdma() and __skb_linearize() calls. Tested on tg3 , bnx2, bonding, this worked very well. This is a generalization of a patch provided by Yi Zou & Jeff Kirsher. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 2c7934f8cf3e..e0c0b86f57a1 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5058,10 +5058,11 @@ int register_netdevice(struct net_device *dev) if (dev->features & NETIF_F_SG) dev->features |= NETIF_F_GSO; - /* Enable GRO for vlans by default if dev->features has GRO also. - * vlan_dev_init() will do the dev->features check. + /* Enable GRO and NETIF_F_HIGHDMA for vlans by default, + * vlan_dev_init() will do the dev->features check, so these features + * are enabled only if supported by underlying device. */ - dev->vlan_features |= NETIF_F_GRO; + dev->vlan_features |= (NETIF_F_GRO | NETIF_F_HIGHDMA); ret = call_netdevice_notifiers(NETDEV_POST_INIT, dev); ret = notifier_to_errno(ret); -- cgit v1.2.3 From 1b4bf461f05d56ced6d6b8f3b4831adc7076f565 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 23 Sep 2010 17:26:35 +0000 Subject: rps: allocate rx queues in register_netdevice only Instead of having two places were we allocate dev->_rx, introduce netif_alloc_rx_queues() helper and call it only from register_netdevice(), not from alloc_netdev_mq() Goal is to let drivers change dev->num_rx_queues after allocating netdev and before registering it. This also removes a lot of ifdefs in net/core/dev.c Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 76 +++++++++++++++++++++++++--------------------------------- 1 file changed, 32 insertions(+), 44 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index e0c0b86f57a1..72e99835e5b8 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4964,6 +4964,34 @@ void netif_stacked_transfer_operstate(const struct net_device *rootdev, } EXPORT_SYMBOL(netif_stacked_transfer_operstate); +static int netif_alloc_rx_queues(struct net_device *dev) +{ +#ifdef CONFIG_RPS + unsigned int i, count = dev->num_rx_queues; + + if (count) { + struct netdev_rx_queue *rx; + + rx = kcalloc(count, sizeof(struct netdev_rx_queue), GFP_KERNEL); + if (!rx) { + pr_err("netdev: Unable to allocate %u rx queues.\n", + count); + return -ENOMEM; + } + dev->_rx = rx; + atomic_set(&rx->count, count); + + /* + * Set a pointer to first element in the array which holds the + * reference count. + */ + for (i = 0; i < count; i++) + rx[i].first = rx; + } +#endif + return 0; +} + /** * register_netdevice - register a network device * @dev: device to register @@ -5001,24 +5029,10 @@ int register_netdevice(struct net_device *dev) dev->iflink = -1; -#ifdef CONFIG_RPS - if (!dev->num_rx_queues) { - /* - * Allocate a single RX queue if driver never called - * alloc_netdev_mq - */ - - dev->_rx = kzalloc(sizeof(struct netdev_rx_queue), GFP_KERNEL); - if (!dev->_rx) { - ret = -ENOMEM; - goto out; - } + ret = netif_alloc_rx_queues(dev); + if (ret) + goto out; - dev->_rx->first = dev->_rx; - atomic_set(&dev->_rx->count, 1); - dev->num_rx_queues = 1; - } -#endif /* Init, if this function is available */ if (dev->netdev_ops->ndo_init) { ret = dev->netdev_ops->ndo_init(dev); @@ -5415,10 +5429,6 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, struct net_device *dev; size_t alloc_size; struct net_device *p; -#ifdef CONFIG_RPS - struct netdev_rx_queue *rx; - int i; -#endif BUG_ON(strlen(name) >= sizeof(dev->name)); @@ -5444,29 +5454,12 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, goto free_p; } -#ifdef CONFIG_RPS - rx = kcalloc(queue_count, sizeof(struct netdev_rx_queue), GFP_KERNEL); - if (!rx) { - printk(KERN_ERR "alloc_netdev: Unable to allocate " - "rx queues.\n"); - goto free_tx; - } - - atomic_set(&rx->count, queue_count); - - /* - * Set a pointer to first element in the array which holds the - * reference count. - */ - for (i = 0; i < queue_count; i++) - rx[i].first = rx; -#endif dev = PTR_ALIGN(p, NETDEV_ALIGN); dev->padded = (char *)dev - (char *)p; if (dev_addr_init(dev)) - goto free_rx; + goto free_tx; dev_mc_init(dev); dev_uc_init(dev); @@ -5478,7 +5471,6 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, dev->real_num_tx_queues = queue_count; #ifdef CONFIG_RPS - dev->_rx = rx; dev->num_rx_queues = queue_count; #endif @@ -5496,11 +5488,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, strcpy(dev->name, name); return dev; -free_rx: -#ifdef CONFIG_RPS - kfree(rx); free_tx: -#endif kfree(tx); free_p: kfree(p); -- cgit v1.2.3 From 7fa7cb7109d07c29ab28bb877bc7049a0150dbe5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Sep 2010 04:18:27 +0000 Subject: fib: use atomic_inc_not_zero() in fib_rules_lookup It seems we dont use appropriate refcount increment in an rcu_read_lock() protected section. fib_rule_get() might increment a null refcount and bad things could happen. While fib_nl_delrule() respects an rcu grace period before calling fib_rule_put(), fib_rules_cleanup_ops() calls fib_rule_put() without a grace period. Note : after this patch, we might avoid the synchronize_rcu() call done in fib_nl_delrule() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/fib_rules.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net/core') diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 42e84e08a1be..d0787284cb07 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -225,9 +225,11 @@ jumped: err = ops->action(rule, fl, flags, arg); if (err != -EAGAIN) { - fib_rule_get(rule); - arg->rule = rule; - goto out; + if (likely(atomic_inc_not_zero(&rule->refcnt))) { + arg->rule = rule; + goto out; + } + break; } } -- cgit v1.2.3 From f91ff5b9ff529be8aac2039af63b2c8ea6cd6ebe Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Sep 2010 06:07:30 +0000 Subject: net: sk_{detach|attach}_filter() rcu fixes sk_attach_filter() and sk_detach_filter() are run with socket locked. Use the appropriate rcu_dereference_protected() instead of blocking BH, and rcu_dereference_bh(). There is no point adding BH prevention and memory barrier. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/filter.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'net/core') diff --git a/net/core/filter.c b/net/core/filter.c index 52b051f82a01..7adf50352918 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -638,10 +638,9 @@ int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk) return err; } - rcu_read_lock_bh(); - old_fp = rcu_dereference_bh(sk->sk_filter); + old_fp = rcu_dereference_protected(sk->sk_filter, + sock_owned_by_user(sk)); rcu_assign_pointer(sk->sk_filter, fp); - rcu_read_unlock_bh(); if (old_fp) sk_filter_delayed_uncharge(sk, old_fp); @@ -654,14 +653,13 @@ int sk_detach_filter(struct sock *sk) int ret = -ENOENT; struct sk_filter *filter; - rcu_read_lock_bh(); - filter = rcu_dereference_bh(sk->sk_filter); + filter = rcu_dereference_protected(sk->sk_filter, + sock_owned_by_user(sk)); if (filter) { rcu_assign_pointer(sk->sk_filter, NULL); sk_filter_delayed_uncharge(sk, filter); ret = 0; } - rcu_read_unlock_bh(); return ret; } EXPORT_SYMBOL_GPL(sk_detach_filter); -- cgit v1.2.3 From 62fe0b40abb3484413800edaef9b087a20059acf Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Mon, 27 Sep 2010 08:24:33 +0000 Subject: net: Allow changing number of RX queues after device allocation For RPS, we create a kobject for each RX queue based on the number of queues passed to alloc_netdev_mq(). However, drivers generally do not determine the numbers of hardware queues to use until much later, so this usually represents the maximum number the driver may use and not the actual number in use. For TX queues, drivers can update the actual number using netif_set_real_num_tx_queues(). Add a corresponding function for RX queues, netif_set_real_num_rx_queues(). Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- include/linux/netdevice.h | 16 +++++++++++++++- net/core/dev.c | 45 +++++++++++++++++++++++++++++++++++++++++---- net/core/net-sysfs.c | 32 ++++++++++++++++++-------------- net/core/net-sysfs.h | 4 ++++ 4 files changed, 78 insertions(+), 19 deletions(-) (limited to 'net/core') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 83de0eb7a071..b15732e22eee 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -976,8 +976,11 @@ struct net_device { struct netdev_rx_queue *_rx; - /* Number of RX queues allocated at alloc_netdev_mq() time */ + /* Number of RX queues allocated at register_netdev() time */ unsigned int num_rx_queues; + + /* Number of RX queues currently active in device */ + unsigned int real_num_rx_queues; #endif rx_handler_func_t *rx_handler; @@ -1685,6 +1688,17 @@ static inline int netif_is_multiqueue(const struct net_device *dev) extern void netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq); +#ifdef CONFIG_RPS +extern int netif_set_real_num_rx_queues(struct net_device *dev, + unsigned int rxq); +#else +static inline int netif_set_real_num_rx_queues(struct net_device *dev, + unsigned int rxq) +{ + return 0; +} +#endif + /* Use this variant when it is known for sure that it * is executing from hardware interrupt context or with hardware interrupts * disabled. diff --git a/net/core/dev.c b/net/core/dev.c index 42b200fdf12e..48ad47f402ad 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1567,6 +1567,41 @@ void netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq) } EXPORT_SYMBOL(netif_set_real_num_tx_queues); +#ifdef CONFIG_RPS +/** + * netif_set_real_num_rx_queues - set actual number of RX queues used + * @dev: Network device + * @rxq: Actual number of RX queues + * + * This must be called either with the rtnl_lock held or before + * registration of the net device. Returns 0 on success, or a + * negative error code. If called before registration, it also + * sets the maximum number of queues, and always succeeds. + */ +int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq) +{ + int rc; + + if (dev->reg_state == NETREG_REGISTERED) { + ASSERT_RTNL(); + + if (rxq > dev->num_rx_queues) + return -EINVAL; + + rc = net_rx_queue_update_kobjects(dev, dev->real_num_rx_queues, + rxq); + if (rc) + return rc; + } else { + dev->num_rx_queues = rxq; + } + + dev->real_num_rx_queues = rxq; + return 0; +} +EXPORT_SYMBOL(netif_set_real_num_rx_queues); +#endif + static inline void __netif_reschedule(struct Qdisc *q) { struct softnet_data *sd; @@ -2352,10 +2387,11 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, if (skb_rx_queue_recorded(skb)) { u16 index = skb_get_rx_queue(skb); - if (unlikely(index >= dev->num_rx_queues)) { - WARN_ONCE(dev->num_rx_queues > 1, "%s received packet " - "on queue %u, but number of RX queues is %u\n", - dev->name, index, dev->num_rx_queues); + if (unlikely(index >= dev->real_num_rx_queues)) { + WARN_ONCE(dev->real_num_rx_queues > 1, + "%s received packet on queue %u, but number " + "of RX queues is %u\n", + dev->name, index, dev->real_num_rx_queues); goto done; } rxqueue = dev->_rx + index; @@ -5472,6 +5508,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, #ifdef CONFIG_RPS dev->num_rx_queues = queue_count; + dev->real_num_rx_queues = queue_count; #endif dev->gso_max_size = GSO_MAX_SIZE; diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 76485a3f910b..fa81fd0a488f 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -742,34 +742,38 @@ static int rx_queue_add_kobject(struct net_device *net, int index) return error; } -static int rx_queue_register_kobjects(struct net_device *net) +int +net_rx_queue_update_kobjects(struct net_device *net, int old_num, int new_num) { int i; int error = 0; - net->queues_kset = kset_create_and_add("queues", - NULL, &net->dev.kobj); - if (!net->queues_kset) - return -ENOMEM; - for (i = 0; i < net->num_rx_queues; i++) { + for (i = old_num; i < new_num; i++) { error = rx_queue_add_kobject(net, i); - if (error) + if (error) { + new_num = old_num; break; + } } - if (error) - while (--i >= 0) - kobject_put(&net->_rx[i].kobj); + while (--i >= new_num) + kobject_put(&net->_rx[i].kobj); return error; } -static void rx_queue_remove_kobjects(struct net_device *net) +static int rx_queue_register_kobjects(struct net_device *net) { - int i; + net->queues_kset = kset_create_and_add("queues", + NULL, &net->dev.kobj); + if (!net->queues_kset) + return -ENOMEM; + return net_rx_queue_update_kobjects(net, 0, net->real_num_rx_queues); +} - for (i = 0; i < net->num_rx_queues; i++) - kobject_put(&net->_rx[i].kobj); +static void rx_queue_remove_kobjects(struct net_device *net) +{ + net_rx_queue_update_kobjects(net, net->real_num_rx_queues, 0); kset_unregister(net->queues_kset); } #endif /* CONFIG_RPS */ diff --git a/net/core/net-sysfs.h b/net/core/net-sysfs.h index 805555e8b187..778e1571548d 100644 --- a/net/core/net-sysfs.h +++ b/net/core/net-sysfs.h @@ -4,4 +4,8 @@ int netdev_kobject_init(void); int netdev_register_kobject(struct net_device *); void netdev_unregister_kobject(struct net_device *); +#ifdef CONFIG_RPS +int net_rx_queue_update_kobjects(struct net_device *, int old_num, int new_num); +#endif + #endif -- cgit v1.2.3 From 4465b469008bc03b98a1b8df4e9ae501b6c69d4b Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Sun, 23 May 2010 19:54:12 +0000 Subject: ipv4: Allow configuring subnets as local addresses This patch allows a host to be configured to respond to any address in a specified range as if it were local, without actually needing to configure the address on an interface. This is done through routing table configuration. For instance, to configure a host to respond to any address in 10.1/16 received on eth0 as a local address we can do: ip rule add from all iif eth0 lookup 200 ip route add local 10.1/16 dev lo proto kernel scope host src 127.0.0.1 table 200 This host is now reachable by any 10.1/16 address (route lookup on input for packets received on eth0 can find the route). On output, the rule will not be matched so that this host can still send packets to 10.1/16 (not sent on loopback). Presumably, external routing can be configured to make sense out of this. To make this work, we needed to modify the logic in finding the interface which is assigned a given source address for output (dev_ip_find). We perform a normal fib_lookup instead of just a lookup on the local table, and in the lookup we ignore the input interface for matching. This patch is useful to implement IP-anycast for subnets of virtual addresses. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/net/flow.h | 1 + net/core/fib_rules.c | 3 ++- net/ipv4/fib_frontend.c | 7 +++---- 3 files changed, 6 insertions(+), 5 deletions(-) (limited to 'net/core') diff --git a/include/net/flow.h b/include/net/flow.h index bb08692a20b0..0ac3fb5e0973 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -49,6 +49,7 @@ struct flowi { __u8 proto; __u8 flags; #define FLOWI_FLAG_ANYSRC 0x01 +#define FLOWI_FLAG_MATCH_ANY_IIF 0x02 union { struct { __be16 sport; diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index d0787284cb07..332c2e31d048 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -182,7 +182,8 @@ static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, { int ret = 0; - if (rule->iifindex && (rule->iifindex != fl->iif)) + if (rule->iifindex && (rule->iifindex != fl->iif) && + !(fl->flags & FLOWI_FLAG_MATCH_ANY_IIF)) goto out; if (rule->oifindex && (rule->oifindex != fl->oif)) diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 7d02a9f999fa..981f3c59b334 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -153,17 +153,16 @@ static void fib_flush(struct net *net) struct net_device * ip_dev_find(struct net *net, __be32 addr) { - struct flowi fl = { .nl_u = { .ip4_u = { .daddr = addr } } }; + struct flowi fl = { .nl_u = { .ip4_u = { .daddr = addr } }, + .flags = FLOWI_FLAG_MATCH_ANY_IIF }; struct fib_result res; struct net_device *dev = NULL; - struct fib_table *local_table; #ifdef CONFIG_IP_MULTIPLE_TABLES res.r = NULL; #endif - local_table = fib_get_table(net, RT_TABLE_LOCAL); - if (!local_table || fib_table_lookup(local_table, &fl, &res)) + if (fib_lookup(net, &fl, &res)) return NULL; if (res.type != RTN_LOCAL) goto out; -- cgit v1.2.3 From 745e20f1b626b1be4b100af5d4bf7b3439392f8f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 29 Sep 2010 13:23:09 -0700 Subject: net: add a recursion limit in xmit path As tunnel devices are going to be lockless, we need to make sure a misconfigured machine wont enter an infinite loop. Add a percpu variable, and limit to three the number of stacked xmits. Reported-by: Jesse Gross Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 48ad47f402ad..50daccad6a53 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2177,6 +2177,9 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, return rc; } +static DEFINE_PER_CPU(int, xmit_recursion); +#define RECURSION_LIMIT 3 + /** * dev_queue_xmit - transmit a buffer * @skb: buffer to transmit @@ -2242,10 +2245,15 @@ int dev_queue_xmit(struct sk_buff *skb) if (txq->xmit_lock_owner != cpu) { + if (__this_cpu_read(xmit_recursion) > RECURSION_LIMIT) + goto recursion_alert; + HARD_TX_LOCK(dev, txq, cpu); if (!netif_tx_queue_stopped(txq)) { + __this_cpu_inc(xmit_recursion); rc = dev_hard_start_xmit(skb, dev, txq); + __this_cpu_dec(xmit_recursion); if (dev_xmit_complete(rc)) { HARD_TX_UNLOCK(dev, txq); goto out; @@ -2257,7 +2265,9 @@ int dev_queue_xmit(struct sk_buff *skb) "queue packet!\n", dev->name); } else { /* Recursion is detected! It is possible, - * unfortunately */ + * unfortunately + */ +recursion_alert: if (net_ratelimit()) printk(KERN_CRIT "Dead loop on virtual device " "%s, fix it urgently!\n", dev->name); -- cgit v1.2.3 From bfa5ae63b823f4ffd3483a05f60a93a4a7b7d680 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 28 Sep 2010 05:58:37 +0000 Subject: net: rename netdev rx_queue to ingress_queue There is some confusion with rx_queue name after RPS, and net drivers private rx_queue fields. I suggest to rename "struct net_device"->rx_queue to ingress_queue. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- net/core/dev.c | 8 ++++---- net/sched/sch_api.c | 14 +++++++------- net/sched/sch_generic.c | 8 ++++---- 4 files changed, 16 insertions(+), 16 deletions(-) (limited to 'net/core') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 6f0845e0b888..ceed3474014a 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -986,7 +986,7 @@ struct net_device { rx_handler_func_t *rx_handler; void *rx_handler_data; - struct netdev_queue rx_queue; /* use two cache lines */ + struct netdev_queue ingress_queue; /* use two cache lines */ /* * Cache lines mostly used on transmit path diff --git a/net/core/dev.c b/net/core/dev.c index 50daccad6a53..a313bab1b754 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2720,7 +2720,7 @@ static int ing_filter(struct sk_buff *skb) skb->tc_verd = SET_TC_RTTL(skb->tc_verd, ttl); skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_INGRESS); - rxq = &dev->rx_queue; + rxq = &dev->ingress_queue; q = rxq->qdisc; if (q != &noop_qdisc) { @@ -2737,7 +2737,7 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, struct net_device *orig_dev) { - if (skb->dev->rx_queue.qdisc == &noop_qdisc) + if (skb->dev->ingress_queue.qdisc == &noop_qdisc) goto out; if (*pt_prev) { @@ -4940,7 +4940,7 @@ static void __netdev_init_queue_locks_one(struct net_device *dev, static void netdev_init_queue_locks(struct net_device *dev) { netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL); - __netdev_init_queue_locks_one(dev, &dev->rx_queue, NULL); + __netdev_init_queue_locks_one(dev, &dev->ingress_queue, NULL); } unsigned long netdev_fix_features(unsigned long features, const char *name) @@ -5452,7 +5452,7 @@ static void netdev_init_one_queue(struct net_device *dev, static void netdev_init_queues(struct net_device *dev) { - netdev_init_one_queue(dev, &dev->rx_queue, NULL); + netdev_init_one_queue(dev, &dev->ingress_queue, NULL); netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL); spin_lock_init(&dev->tx_global_lock); } diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 6fb3d41c0e41..b8020784d0e9 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -240,7 +240,7 @@ struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) if (q) goto out; - q = qdisc_match_from_root(dev->rx_queue.qdisc_sleeping, handle); + q = qdisc_match_from_root(dev->ingress_queue.qdisc_sleeping, handle); out: return q; } @@ -701,7 +701,7 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, } for (i = 0; i < num_q; i++) { - struct netdev_queue *dev_queue = &dev->rx_queue; + struct netdev_queue *dev_queue = &dev->ingress_queue; if (!ingress) dev_queue = netdev_get_tx_queue(dev, i); @@ -979,7 +979,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg) return -ENOENT; q = qdisc_leaf(p, clid); } else { /* ingress */ - q = dev->rx_queue.qdisc_sleeping; + q = dev->ingress_queue.qdisc_sleeping; } } else { q = dev->qdisc; @@ -1044,7 +1044,7 @@ replay: return -ENOENT; q = qdisc_leaf(p, clid); } else { /*ingress */ - q = dev->rx_queue.qdisc_sleeping; + q = dev->ingress_queue.qdisc_sleeping; } } else { q = dev->qdisc; @@ -1124,7 +1124,7 @@ create_n_graft: if (!(n->nlmsg_flags&NLM_F_CREATE)) return -ENOENT; if (clid == TC_H_INGRESS) - q = qdisc_create(dev, &dev->rx_queue, p, + q = qdisc_create(dev, &dev->ingress_queue, p, tcm->tcm_parent, tcm->tcm_parent, tca, &err); else { @@ -1304,7 +1304,7 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb) if (tc_dump_qdisc_root(dev->qdisc, skb, cb, &q_idx, s_q_idx) < 0) goto done; - dev_queue = &dev->rx_queue; + dev_queue = &dev->ingress_queue; if (tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb, &q_idx, s_q_idx) < 0) goto done; @@ -1595,7 +1595,7 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb) if (tc_dump_tclass_root(dev->qdisc, skb, tcm, cb, &t, s_t) < 0) goto done; - dev_queue = &dev->rx_queue; + dev_queue = &dev->ingress_queue; if (tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, &t, s_t) < 0) goto done; diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 2aeb3a4386a1..545278a1c478 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -753,7 +753,7 @@ void dev_activate(struct net_device *dev) need_watchdog = 0; netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog); - transition_one_qdisc(dev, &dev->rx_queue, NULL); + transition_one_qdisc(dev, &dev->ingress_queue, NULL); if (need_watchdog) { dev->trans_start = jiffies; @@ -812,7 +812,7 @@ static bool some_qdisc_is_busy(struct net_device *dev) void dev_deactivate(struct net_device *dev) { netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc); - dev_deactivate_queue(dev, &dev->rx_queue, &noop_qdisc); + dev_deactivate_queue(dev, &dev->ingress_queue, &noop_qdisc); dev_watchdog_down(dev); @@ -838,7 +838,7 @@ void dev_init_scheduler(struct net_device *dev) { dev->qdisc = &noop_qdisc; netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc); - dev_init_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc); + dev_init_scheduler_queue(dev, &dev->ingress_queue, &noop_qdisc); setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev); } @@ -861,7 +861,7 @@ static void shutdown_scheduler_queue(struct net_device *dev, void dev_shutdown(struct net_device *dev) { netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc); - shutdown_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc); + shutdown_scheduler_queue(dev, &dev->ingress_queue, &noop_qdisc); qdisc_destroy(dev->qdisc); dev->qdisc = &noop_qdisc; -- cgit v1.2.3 From c7d4426a98a5f6654cd0b4b33d9dab2e77192c18 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 3 Oct 2010 22:17:54 -0700 Subject: net: introduce DST_NOCACHE flag While doing stress tests with IP route cache disabled, and multi queue devices, I noticed a very high contention on one rwlock used in neighbour code. When many cpus are trying to send frames (possibly using a high performance multiqueue device) to the same neighbour, they fight for the neigh->lock rwlock in order to call neigh_hh_init(), and fight on hh->hh_refcnt (a pair of atomic_inc/atomic_dec_and_test()) But we dont need to call neigh_hh_init() for dst that are used only once. It costs four atomic operations at least, on two contended cache lines, plus the high contention on neigh->lock rwlock. Introduce a new dst flag, DST_NOCACHE, that is set when dst was not inserted in route cache. With the stress test bench, sending 160000000 frames on one neighbour, results are : Before patch: real 2m28.406s user 0m11.781s sys 36m17.964s After patch: real 1m26.532s user 0m12.185s sys 20m3.903s Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/dst.h | 9 +++++---- net/core/neighbour.c | 4 +++- net/ipv4/route.c | 1 + 3 files changed, 9 insertions(+), 5 deletions(-) (limited to 'net/core') diff --git a/include/net/dst.h b/include/net/dst.h index aa53fbc34b2b..a217c838ec0d 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -43,10 +43,11 @@ struct dst_entry { short error; short obsolete; int flags; -#define DST_HOST 1 -#define DST_NOXFRM 2 -#define DST_NOPOLICY 4 -#define DST_NOHASH 8 +#define DST_HOST 0x0001 +#define DST_NOXFRM 0x0002 +#define DST_NOPOLICY 0x0004 +#define DST_NOHASH 0x0008 +#define DST_NOCACHE 0x0010 unsigned long expires; unsigned short header_len; /* more space at head required */ diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 96b1a749abb4..b142a0d76072 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1210,7 +1210,9 @@ int neigh_resolve_output(struct sk_buff *skb) if (!neigh_event_send(neigh, skb)) { int err; struct net_device *dev = neigh->dev; - if (dev->header_ops->cache && !dst->hh) { + if (dev->header_ops->cache && + !dst->hh && + !(dst->flags & DST_NOCACHE)) { write_lock_bh(&neigh->lock); if (!dst->hh) neigh_hh_init(neigh, dst, dst->ops->protocol); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index a61acea975f1..c3cb8bd23638 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1107,6 +1107,7 @@ restart: * on the route gc list. */ + rt->dst.flags |= DST_NOCACHE; if (rt->rt_type == RTN_UNICAST || rt->fl.iif == 0) { int err = arp_bind_neighbour(&rt->dst); if (err) { -- cgit v1.2.3 From 24824a09e35402b8d58dcc5be803a5ad3937bdba Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 2 Oct 2010 06:11:55 +0000 Subject: net: dynamic ingress_queue allocation ingress being not used very much, and net_device->ingress_queue being quite a big object (128 or 256 bytes), use a dynamic allocation if needed (tc qdisc add dev eth0 ingress ...) dev_ingress_queue(dev) helper should be used only with RTNL taken. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- include/linux/rtnetlink.h | 8 ++++++++ net/core/dev.c | 34 ++++++++++++++++++++++++++-------- net/sched/sch_api.c | 42 ++++++++++++++++++++++++++++-------------- net/sched/sch_generic.c | 12 ++++++++---- 5 files changed, 71 insertions(+), 27 deletions(-) (limited to 'net/core') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ceed3474014a..92d81edd5808 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -986,7 +986,7 @@ struct net_device { rx_handler_func_t *rx_handler; void *rx_handler_data; - struct netdev_queue ingress_queue; /* use two cache lines */ + struct netdev_queue __rcu *ingress_queue; /* * Cache lines mostly used on transmit path diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 68c436bddc88..0bb7b48632bd 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -6,6 +6,7 @@ #include #include #include +#include /* rtnetlink families. Values up to 127 are reserved for real address * families, values above 128 may be used arbitrarily. @@ -769,6 +770,13 @@ extern int lockdep_rtnl_is_held(void); #define rtnl_dereference(p) \ rcu_dereference_check(p, lockdep_rtnl_is_held()) +static inline struct netdev_queue *dev_ingress_queue(struct net_device *dev) +{ + return rtnl_dereference(dev->ingress_queue); +} + +extern struct netdev_queue *dev_ingress_queue_create(struct net_device *dev); + extern void rtnetlink_init(void); extern void __rtnl_unlock(void); diff --git a/net/core/dev.c b/net/core/dev.c index a313bab1b754..ce6ad88c980b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2702,11 +2702,10 @@ EXPORT_SYMBOL_GPL(br_fdb_test_addr_hook); * the ingress scheduler, you just cant add policies on ingress. * */ -static int ing_filter(struct sk_buff *skb) +static int ing_filter(struct sk_buff *skb, struct netdev_queue *rxq) { struct net_device *dev = skb->dev; u32 ttl = G_TC_RTTL(skb->tc_verd); - struct netdev_queue *rxq; int result = TC_ACT_OK; struct Qdisc *q; @@ -2720,8 +2719,6 @@ static int ing_filter(struct sk_buff *skb) skb->tc_verd = SET_TC_RTTL(skb->tc_verd, ttl); skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_INGRESS); - rxq = &dev->ingress_queue; - q = rxq->qdisc; if (q != &noop_qdisc) { spin_lock(qdisc_lock(q)); @@ -2737,7 +2734,9 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, struct net_device *orig_dev) { - if (skb->dev->ingress_queue.qdisc == &noop_qdisc) + struct netdev_queue *rxq = rcu_dereference(skb->dev->ingress_queue); + + if (!rxq || rxq->qdisc == &noop_qdisc) goto out; if (*pt_prev) { @@ -2745,7 +2744,7 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb, *pt_prev = NULL; } - switch (ing_filter(skb)) { + switch (ing_filter(skb, rxq)) { case TC_ACT_SHOT: case TC_ACT_STOLEN: kfree_skb(skb); @@ -4940,7 +4939,6 @@ static void __netdev_init_queue_locks_one(struct net_device *dev, static void netdev_init_queue_locks(struct net_device *dev) { netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL); - __netdev_init_queue_locks_one(dev, &dev->ingress_queue, NULL); } unsigned long netdev_fix_features(unsigned long features, const char *name) @@ -5452,11 +5450,29 @@ static void netdev_init_one_queue(struct net_device *dev, static void netdev_init_queues(struct net_device *dev) { - netdev_init_one_queue(dev, &dev->ingress_queue, NULL); netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL); spin_lock_init(&dev->tx_global_lock); } +struct netdev_queue *dev_ingress_queue_create(struct net_device *dev) +{ + struct netdev_queue *queue = dev_ingress_queue(dev); + +#ifdef CONFIG_NET_CLS_ACT + if (queue) + return queue; + queue = kzalloc(sizeof(*queue), GFP_KERNEL); + if (!queue) + return NULL; + netdev_init_one_queue(dev, queue, NULL); + __netdev_init_queue_locks_one(dev, queue, NULL); + queue->qdisc = &noop_qdisc; + queue->qdisc_sleeping = &noop_qdisc; + rcu_assign_pointer(dev->ingress_queue, queue); +#endif + return queue; +} + /** * alloc_netdev_mq - allocate network device * @sizeof_priv: size of private data to allocate space for @@ -5559,6 +5575,8 @@ void free_netdev(struct net_device *dev) kfree(dev->_tx); + kfree(rcu_dereference_raw(dev->ingress_queue)); + /* Flush device addresses */ dev_addr_flush(dev); diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index b8020784d0e9..b22ca2d1cebc 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -240,7 +240,10 @@ struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) if (q) goto out; - q = qdisc_match_from_root(dev->ingress_queue.qdisc_sleeping, handle); + if (dev_ingress_queue(dev)) + q = qdisc_match_from_root( + dev_ingress_queue(dev)->qdisc_sleeping, + handle); out: return q; } @@ -690,6 +693,8 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, (new && new->flags & TCQ_F_INGRESS)) { num_q = 1; ingress = 1; + if (!dev_ingress_queue(dev)) + return -ENOENT; } if (dev->flags & IFF_UP) @@ -701,7 +706,7 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, } for (i = 0; i < num_q; i++) { - struct netdev_queue *dev_queue = &dev->ingress_queue; + struct netdev_queue *dev_queue = dev_ingress_queue(dev); if (!ingress) dev_queue = netdev_get_tx_queue(dev, i); @@ -979,7 +984,8 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg) return -ENOENT; q = qdisc_leaf(p, clid); } else { /* ingress */ - q = dev->ingress_queue.qdisc_sleeping; + if (dev_ingress_queue(dev)) + q = dev_ingress_queue(dev)->qdisc_sleeping; } } else { q = dev->qdisc; @@ -1043,8 +1049,9 @@ replay: if ((p = qdisc_lookup(dev, TC_H_MAJ(clid))) == NULL) return -ENOENT; q = qdisc_leaf(p, clid); - } else { /*ingress */ - q = dev->ingress_queue.qdisc_sleeping; + } else { /* ingress */ + if (dev_ingress_queue_create(dev)) + q = dev_ingress_queue(dev)->qdisc_sleeping; } } else { q = dev->qdisc; @@ -1123,11 +1130,14 @@ replay: create_n_graft: if (!(n->nlmsg_flags&NLM_F_CREATE)) return -ENOENT; - if (clid == TC_H_INGRESS) - q = qdisc_create(dev, &dev->ingress_queue, p, - tcm->tcm_parent, tcm->tcm_parent, - tca, &err); - else { + if (clid == TC_H_INGRESS) { + if (dev_ingress_queue(dev)) + q = qdisc_create(dev, dev_ingress_queue(dev), p, + tcm->tcm_parent, tcm->tcm_parent, + tca, &err); + else + err = -ENOENT; + } else { struct netdev_queue *dev_queue; if (p && p->ops->cl_ops && p->ops->cl_ops->select_queue) @@ -1304,8 +1314,10 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb) if (tc_dump_qdisc_root(dev->qdisc, skb, cb, &q_idx, s_q_idx) < 0) goto done; - dev_queue = &dev->ingress_queue; - if (tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb, &q_idx, s_q_idx) < 0) + dev_queue = dev_ingress_queue(dev); + if (dev_queue && + tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb, + &q_idx, s_q_idx) < 0) goto done; cont: @@ -1595,8 +1607,10 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb) if (tc_dump_tclass_root(dev->qdisc, skb, tcm, cb, &t, s_t) < 0) goto done; - dev_queue = &dev->ingress_queue; - if (tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, &t, s_t) < 0) + dev_queue = dev_ingress_queue(dev); + if (dev_queue && + tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, + &t, s_t) < 0) goto done; done: diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 545278a1c478..3d57681bdb76 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -753,7 +753,8 @@ void dev_activate(struct net_device *dev) need_watchdog = 0; netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog); - transition_one_qdisc(dev, &dev->ingress_queue, NULL); + if (dev_ingress_queue(dev)) + transition_one_qdisc(dev, dev_ingress_queue(dev), NULL); if (need_watchdog) { dev->trans_start = jiffies; @@ -812,7 +813,8 @@ static bool some_qdisc_is_busy(struct net_device *dev) void dev_deactivate(struct net_device *dev) { netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc); - dev_deactivate_queue(dev, &dev->ingress_queue, &noop_qdisc); + if (dev_ingress_queue(dev)) + dev_deactivate_queue(dev, dev_ingress_queue(dev), &noop_qdisc); dev_watchdog_down(dev); @@ -838,7 +840,8 @@ void dev_init_scheduler(struct net_device *dev) { dev->qdisc = &noop_qdisc; netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc); - dev_init_scheduler_queue(dev, &dev->ingress_queue, &noop_qdisc); + if (dev_ingress_queue(dev)) + dev_init_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc); setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev); } @@ -861,7 +864,8 @@ static void shutdown_scheduler_queue(struct net_device *dev, void dev_shutdown(struct net_device *dev) { netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc); - shutdown_scheduler_queue(dev, &dev->ingress_queue, &noop_qdisc); + if (dev_ingress_queue(dev)) + shutdown_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc); qdisc_destroy(dev->qdisc); dev->qdisc = &noop_qdisc; -- cgit v1.2.3 From 1df9916e46451533463f227e6be57cc2cfca4c5f Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Mon, 4 Oct 2010 20:14:17 +0000 Subject: fib: fib_rules_cleanup can be static fib_rules_cleanup_ups is only defined and used in one place. Signed-off-by: Stephen Hemminger Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/fib_rules.h | 1 - net/core/fib_rules.c | 3 +-- 2 files changed, 1 insertion(+), 3 deletions(-) (limited to 'net/core') diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index e8923bc20f9f..ac2fd002812e 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -106,7 +106,6 @@ static inline u32 frh_get_table(struct fib_rule_hdr *frh, struct nlattr **nla) extern struct fib_rules_ops *fib_rules_register(const struct fib_rules_ops *, struct net *); extern void fib_rules_unregister(struct fib_rules_ops *); -extern void fib_rules_cleanup_ops(struct fib_rules_ops *); extern int fib_rules_lookup(struct fib_rules_ops *, struct flowi *, int flags, diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 332c2e31d048..cfb7d25c172d 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -144,7 +144,7 @@ fib_rules_register(const struct fib_rules_ops *tmpl, struct net *net) } EXPORT_SYMBOL_GPL(fib_rules_register); -void fib_rules_cleanup_ops(struct fib_rules_ops *ops) +static void fib_rules_cleanup_ops(struct fib_rules_ops *ops) { struct fib_rule *rule, *tmp; @@ -153,7 +153,6 @@ void fib_rules_cleanup_ops(struct fib_rules_ops *ops) fib_rule_put(rule); } } -EXPORT_SYMBOL_GPL(fib_rules_cleanup_ops); static void fib_rules_put_rcu(struct rcu_head *head) { -- cgit v1.2.3 From caf586e5f23cebb2a68cbaf288d59dbbf2d74052 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 30 Sep 2010 21:06:55 +0000 Subject: net: add a core netdev->rx_dropped counter In various situations, a device provides a packet to our stack and we drop it before it enters protocol stack : - softnet backlog full (accounted in /proc/net/softnet_stat) - bad vlan tag (not accounted) - unknown/unregistered protocol (not accounted) We can handle a per-device counter of such dropped frames at core level, and automatically adds it to the device provided stats (rx_dropped), so that standard tools can be used (ifconfig, ip link, cat /proc/net/dev) This is a generalization of commit 8990f468a (net: rx_dropped accounting), thus reverting it. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/loopback.c | 8 +------- include/linux/netdevice.h | 3 +++ net/8021q/vlan.h | 2 -- net/8021q/vlan_core.c | 2 ++ net/8021q/vlan_dev.c | 11 ++++------- net/core/dev.c | 19 +++++++++++-------- net/ipv4/ip_gre.c | 3 +-- net/ipv4/ipip.c | 3 +-- net/ipv6/ip6_tunnel.c | 3 +-- net/ipv6/ip6mr.c | 3 +-- net/ipv6/sit.c | 3 +-- 11 files changed, 26 insertions(+), 34 deletions(-) (limited to 'net/core') diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index 4b0e30b564e5..2d9663a1c54d 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -64,7 +64,6 @@ struct pcpu_lstats { u64 packets; u64 bytes; struct u64_stats_sync syncp; - unsigned long drops; }; /* @@ -90,8 +89,7 @@ static netdev_tx_t loopback_xmit(struct sk_buff *skb, lb_stats->bytes += len; lb_stats->packets++; u64_stats_update_end(&lb_stats->syncp); - } else - lb_stats->drops++; + } return NETDEV_TX_OK; } @@ -101,7 +99,6 @@ static struct rtnl_link_stats64 *loopback_get_stats64(struct net_device *dev, { u64 bytes = 0; u64 packets = 0; - u64 drops = 0; int i; for_each_possible_cpu(i) { @@ -115,14 +112,11 @@ static struct rtnl_link_stats64 *loopback_get_stats64(struct net_device *dev, tbytes = lb_stats->bytes; tpackets = lb_stats->packets; } while (u64_stats_fetch_retry(&lb_stats->syncp, start)); - drops += lb_stats->drops; bytes += tbytes; packets += tpackets; } stats->rx_packets = packets; stats->tx_packets = packets; - stats->rx_dropped = drops; - stats->rx_errors = drops; stats->rx_bytes = bytes; stats->tx_bytes = bytes; return stats; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 92d81edd5808..6abcef67b178 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -884,6 +884,9 @@ struct net_device { int iflink; struct net_device_stats stats; + atomic_long_t rx_dropped; /* dropped packets by core network + * Do not use this in drivers. + */ #ifdef CONFIG_WIRELESS_EXT /* List of functions to handle Wireless Extensions (instead of ioctl). diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h index b26ce343072c..8d9503ad01da 100644 --- a/net/8021q/vlan.h +++ b/net/8021q/vlan.h @@ -25,7 +25,6 @@ struct vlan_priority_tci_mapping { * @rx_multicast: number of received multicast packets * @syncp: synchronization point for 64bit counters * @rx_errors: number of errors - * @rx_dropped: number of dropped packets */ struct vlan_rx_stats { u64 rx_packets; @@ -33,7 +32,6 @@ struct vlan_rx_stats { u64 rx_multicast; struct u64_stats_sync syncp; unsigned long rx_errors; - unsigned long rx_dropped; }; /** diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index b6d55a9304f2..dee727ce0291 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -33,6 +33,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, return polling ? netif_receive_skb(skb) : netif_rx(skb); drop: + atomic_long_inc(&skb->dev->rx_dropped); dev_kfree_skb_any(skb); return NET_RX_DROP; } @@ -123,6 +124,7 @@ vlan_gro_common(struct napi_struct *napi, struct vlan_group *grp, return dev_gro_receive(napi, skb); drop: + atomic_long_inc(&skb->dev->rx_dropped); return GRO_DROP; } diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index f6fbcc0f1af9..f54251edd40d 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -225,16 +225,15 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev, } } - if (unlikely(netif_rx(skb) == NET_RX_DROP)) { - if (rx_stats) - rx_stats->rx_dropped++; - } + netif_rx(skb); + rcu_read_unlock(); return NET_RX_SUCCESS; err_unlock: rcu_read_unlock(); err_free: + atomic_long_inc(&dev->rx_dropped); kfree_skb(skb); return NET_RX_DROP; } @@ -846,15 +845,13 @@ static struct rtnl_link_stats64 *vlan_dev_get_stats64(struct net_device *dev, st accum.rx_packets += rxpackets; accum.rx_bytes += rxbytes; accum.rx_multicast += rxmulticast; - /* rx_errors, rx_dropped are ulong, not protected by syncp */ + /* rx_errors is ulong, not protected by syncp */ accum.rx_errors += p->rx_errors; - accum.rx_dropped += p->rx_dropped; } stats->rx_packets = accum.rx_packets; stats->rx_bytes = accum.rx_bytes; stats->rx_errors = accum.rx_errors; stats->multicast = accum.rx_multicast; - stats->rx_dropped = accum.rx_dropped; } return stats; } diff --git a/net/core/dev.c b/net/core/dev.c index ce6ad88c980b..7d149550e8d6 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1483,8 +1483,9 @@ int dev_forward_skb(struct net_device *dev, struct sk_buff *skb) skb_orphan(skb); nf_reset(skb); - if (!(dev->flags & IFF_UP) || - (skb->len > (dev->mtu + dev->hard_header_len))) { + if (unlikely(!(dev->flags & IFF_UP) || + (skb->len > (dev->mtu + dev->hard_header_len)))) { + atomic_long_inc(&dev->rx_dropped); kfree_skb(skb); return NET_RX_DROP; } @@ -2548,6 +2549,7 @@ enqueue: local_irq_restore(flags); + atomic_long_inc(&skb->dev->rx_dropped); kfree_skb(skb); return NET_RX_DROP; } @@ -2995,6 +2997,7 @@ ncls: if (pt_prev) { ret = pt_prev->func(skb, skb->dev, pt_prev, orig_dev); } else { + atomic_long_inc(&skb->dev->rx_dropped); kfree_skb(skb); /* Jamal, now you will not able to escape explaining * me how you were going to use this. :-) @@ -5429,14 +5432,14 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev, if (ops->ndo_get_stats64) { memset(storage, 0, sizeof(*storage)); - return ops->ndo_get_stats64(dev, storage); - } - if (ops->ndo_get_stats) { + ops->ndo_get_stats64(dev, storage); + } else if (ops->ndo_get_stats) { netdev_stats_to_stats64(storage, ops->ndo_get_stats(dev)); - return storage; + } else { + netdev_stats_to_stats64(storage, &dev->stats); + dev_txq_stats_fold(dev, storage); } - netdev_stats_to_stats64(storage, &dev->stats); - dev_txq_stats_fold(dev, storage); + storage->rx_dropped += atomic_long_read(&dev->rx_dropped); return storage; } EXPORT_SYMBOL(dev_get_stats); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index fbe2c473a06a..9d421f4cf3ef 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -679,8 +679,7 @@ static int ipgre_rcv(struct sk_buff *skb) skb_reset_network_header(skb); ipgre_ecn_decapsulate(iph, skb); - if (netif_rx(skb) == NET_RX_DROP) - tunnel->dev->stats.rx_dropped++; + netif_rx(skb); rcu_read_unlock(); return 0; diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 6ad46c28ede2..e9b816e6cd73 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -414,8 +414,7 @@ static int ipip_rcv(struct sk_buff *skb) ipip_ecn_decapsulate(iph, skb); - if (netif_rx(skb) == NET_RX_DROP) - tunnel->dev->stats.rx_dropped++; + netif_rx(skb); rcu_read_unlock(); return 0; diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 8be3c452af90..c2c0f89397b1 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -768,8 +768,7 @@ static int ip6_tnl_rcv(struct sk_buff *skb, __u16 protocol, dscp_ecn_decapsulate(t, ipv6h, skb); - if (netif_rx(skb) == NET_RX_DROP) - t->dev->stats.rx_dropped++; + netif_rx(skb); rcu_read_unlock(); return 0; diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 2640c9be589d..6f32ffce7022 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -666,8 +666,7 @@ static int pim6_rcv(struct sk_buff *skb) skb_tunnel_rx(skb, reg_dev); - if (netif_rx(skb) == NET_RX_DROP) - reg_dev->stats.rx_dropped++; + netif_rx(skb); dev_put(reg_dev); return 0; diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index d7701782b639..367a6cc584cc 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -600,8 +600,7 @@ static int ipip6_rcv(struct sk_buff *skb) ipip6_ecn_decapsulate(iph, skb); - if (netif_rx(skb) == NET_RX_DROP) - tunnel->dev->stats.rx_dropped++; + netif_rx(skb); rcu_read_unlock(); return 0; -- cgit v1.2.3 From 110b2499370c401cdcc7c63e481084467291d556 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 4 Oct 2010 04:27:36 +0000 Subject: net neigh: neigh_delete() and neigh_add() changes neigh_delete() and neigh_add() dont need to touch device refcount, we hold RTNL when calling them, so device cannot disappear under us. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/neighbour.c | 39 +++++++++++++++++---------------------- 1 file changed, 17 insertions(+), 22 deletions(-) (limited to 'net/core') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index b142a0d76072..d6996e072a41 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1531,6 +1531,7 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) struct net_device *dev = NULL; int err = -EINVAL; + ASSERT_RTNL(); if (nlmsg_len(nlh) < sizeof(*ndm)) goto out; @@ -1540,7 +1541,7 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) ndm = nlmsg_data(nlh); if (ndm->ndm_ifindex) { - dev = dev_get_by_index(net, ndm->ndm_ifindex); + dev = __dev_get_by_index(net, ndm->ndm_ifindex); if (dev == NULL) { err = -ENODEV; goto out; @@ -1556,34 +1557,31 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) read_unlock(&neigh_tbl_lock); if (nla_len(dst_attr) < tbl->key_len) - goto out_dev_put; + goto out; if (ndm->ndm_flags & NTF_PROXY) { err = pneigh_delete(tbl, net, nla_data(dst_attr), dev); - goto out_dev_put; + goto out; } if (dev == NULL) - goto out_dev_put; + goto out; neigh = neigh_lookup(tbl, nla_data(dst_attr), dev); if (neigh == NULL) { err = -ENOENT; - goto out_dev_put; + goto out; } err = neigh_update(neigh, NULL, NUD_FAILED, NEIGH_UPDATE_F_OVERRIDE | NEIGH_UPDATE_F_ADMIN); neigh_release(neigh); - goto out_dev_put; + goto out; } read_unlock(&neigh_tbl_lock); err = -EAFNOSUPPORT; -out_dev_put: - if (dev) - dev_put(dev); out: return err; } @@ -1597,6 +1595,7 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) struct net_device *dev = NULL; int err; + ASSERT_RTNL(); err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL); if (err < 0) goto out; @@ -1607,14 +1606,14 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) ndm = nlmsg_data(nlh); if (ndm->ndm_ifindex) { - dev = dev_get_by_index(net, ndm->ndm_ifindex); + dev = __dev_get_by_index(net, ndm->ndm_ifindex); if (dev == NULL) { err = -ENODEV; goto out; } if (tb[NDA_LLADDR] && nla_len(tb[NDA_LLADDR]) < dev->addr_len) - goto out_dev_put; + goto out; } read_lock(&neigh_tbl_lock); @@ -1628,7 +1627,7 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) read_unlock(&neigh_tbl_lock); if (nla_len(tb[NDA_DST]) < tbl->key_len) - goto out_dev_put; + goto out; dst = nla_data(tb[NDA_DST]); lladdr = tb[NDA_LLADDR] ? nla_data(tb[NDA_LLADDR]) : NULL; @@ -1641,29 +1640,29 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) pn->flags = ndm->ndm_flags; err = 0; } - goto out_dev_put; + goto out; } if (dev == NULL) - goto out_dev_put; + goto out; neigh = neigh_lookup(tbl, dst, dev); if (neigh == NULL) { if (!(nlh->nlmsg_flags & NLM_F_CREATE)) { err = -ENOENT; - goto out_dev_put; + goto out; } neigh = __neigh_lookup_errno(tbl, dst, dev); if (IS_ERR(neigh)) { err = PTR_ERR(neigh); - goto out_dev_put; + goto out; } } else { if (nlh->nlmsg_flags & NLM_F_EXCL) { err = -EEXIST; neigh_release(neigh); - goto out_dev_put; + goto out; } if (!(nlh->nlmsg_flags & NLM_F_REPLACE)) @@ -1676,15 +1675,11 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) } else err = neigh_update(neigh, lladdr, ndm->ndm_state, flags); neigh_release(neigh); - goto out_dev_put; + goto out; } read_unlock(&neigh_tbl_lock); err = -EAFNOSUPPORT; - -out_dev_put: - if (dev) - dev_put(dev); out: return err; } -- cgit v1.2.3 From d6bf781712a1d25cc8987036b3a48535b331eb91 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 4 Oct 2010 06:15:44 +0000 Subject: net neigh: RCU conversion of neigh hash table David This is the first step for RCU conversion of neigh code. Next patches will convert hash_buckets[] and "struct neighbour" to RCU protected objects. Thanks [PATCH net-next] net neigh: RCU conversion of neigh hash table Instead of storing hash_buckets, hash_mask and hash_rnd in "struct neigh_table", a new structure is defined : struct neigh_hash_table { struct neighbour **hash_buckets; unsigned int hash_mask; __u32 hash_rnd; struct rcu_head rcu; }; And "struct neigh_table" has an RCU protected pointer to such a neigh_hash_table. This means the signature of (*hash)() function changed: We need to add a third parameter with the actual hash_rnd value, since this is not anymore a neigh_table field. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/neighbour.h | 16 +++- net/atm/clip.c | 4 +- net/core/neighbour.c | 219 ++++++++++++++++++++++++++++++------------------ net/decnet/dn_neigh.c | 13 +-- net/ipv4/arp.c | 8 +- net/ipv6/ndisc.c | 10 ++- 6 files changed, 170 insertions(+), 100 deletions(-) (limited to 'net/core') diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 7d08fd1062f0..37845dae6488 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -138,13 +138,22 @@ struct pneigh_entry { * neighbour table manipulation */ +struct neigh_hash_table { + struct neighbour **hash_buckets; + unsigned int hash_mask; + __u32 hash_rnd; + struct rcu_head rcu; +}; + struct neigh_table { struct neigh_table *next; int family; int entry_size; int key_len; - __u32 (*hash)(const void *pkey, const struct net_device *); + __u32 (*hash)(const void *pkey, + const struct net_device *dev, + __u32 hash_rnd); int (*constructor)(struct neighbour *); int (*pconstructor)(struct pneigh_entry *); void (*pdestructor)(struct pneigh_entry *); @@ -165,9 +174,7 @@ struct neigh_table { unsigned long last_rand; struct kmem_cache *kmem_cachep; struct neigh_statistics __percpu *stats; - struct neighbour **hash_buckets; - unsigned int hash_mask; - __u32 hash_rnd; + struct neigh_hash_table __rcu *nht; struct pneigh_entry **phash_buckets; }; @@ -237,6 +244,7 @@ extern void pneigh_for_each(struct neigh_table *tbl, void (*cb)(struct pneigh_en struct neigh_seq_state { struct seq_net_private p; struct neigh_table *tbl; + struct neigh_hash_table *nht; void *(*neigh_sub_iter)(struct neigh_seq_state *state, struct neighbour *n, loff_t *pos); unsigned int bucket; diff --git a/net/atm/clip.c b/net/atm/clip.c index 95fdd1185067..ff956d1115bc 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -310,9 +310,9 @@ static int clip_constructor(struct neighbour *neigh) return 0; } -static u32 clip_hash(const void *pkey, const struct net_device *dev) +static u32 clip_hash(const void *pkey, const struct net_device *dev, __u32 rnd) { - return jhash_2words(*(u32 *) pkey, dev->ifindex, clip_tbl.hash_rnd); + return jhash_2words(*(u32 *) pkey, dev->ifindex, rnd); } static struct neigh_table clip_tbl = { diff --git a/net/core/neighbour.c b/net/core/neighbour.c index d6996e072a41..dd8920e4f508 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -131,14 +131,17 @@ static int neigh_forced_gc(struct neigh_table *tbl) { int shrunk = 0; int i; + struct neigh_hash_table *nht; NEIGH_CACHE_STAT_INC(tbl, forced_gc_runs); write_lock_bh(&tbl->lock); - for (i = 0; i <= tbl->hash_mask; i++) { + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); + for (i = 0; i <= nht->hash_mask; i++) { struct neighbour *n, **np; - np = &tbl->hash_buckets[i]; + np = &nht->hash_buckets[i]; while ((n = *np) != NULL) { /* Neighbour record may be discarded if: * - nobody refers to it. @@ -199,9 +202,13 @@ static void pneigh_queue_purge(struct sk_buff_head *list) static void neigh_flush_dev(struct neigh_table *tbl, struct net_device *dev) { int i; + struct neigh_hash_table *nht; - for (i = 0; i <= tbl->hash_mask; i++) { - struct neighbour *n, **np = &tbl->hash_buckets[i]; + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); + + for (i = 0; i <= nht->hash_mask; i++) { + struct neighbour *n, **np = &nht->hash_buckets[i]; while ((n = *np) != NULL) { if (dev && n->dev != dev) { @@ -297,64 +304,81 @@ out_entries: goto out; } -static struct neighbour **neigh_hash_alloc(unsigned int entries) +static struct neigh_hash_table *neigh_hash_alloc(unsigned int entries) { - unsigned long size = entries * sizeof(struct neighbour *); - struct neighbour **ret; + size_t size = entries * sizeof(struct neighbour *); + struct neigh_hash_table *ret; + struct neighbour **buckets; - if (size <= PAGE_SIZE) { - ret = kzalloc(size, GFP_ATOMIC); - } else { - ret = (struct neighbour **) - __get_free_pages(GFP_ATOMIC|__GFP_ZERO, get_order(size)); + ret = kmalloc(sizeof(*ret), GFP_ATOMIC); + if (!ret) + return NULL; + if (size <= PAGE_SIZE) + buckets = kzalloc(size, GFP_ATOMIC); + else + buckets = (struct neighbour **) + __get_free_pages(GFP_ATOMIC | __GFP_ZERO, + get_order(size)); + if (!buckets) { + kfree(ret); + return NULL; } + ret->hash_buckets = buckets; + ret->hash_mask = entries - 1; + get_random_bytes(&ret->hash_rnd, sizeof(ret->hash_rnd)); return ret; } -static void neigh_hash_free(struct neighbour **hash, unsigned int entries) +static void neigh_hash_free_rcu(struct rcu_head *head) { - unsigned long size = entries * sizeof(struct neighbour *); + struct neigh_hash_table *nht = container_of(head, + struct neigh_hash_table, + rcu); + size_t size = (nht->hash_mask + 1) * sizeof(struct neighbour *); + struct neighbour **buckets = nht->hash_buckets; if (size <= PAGE_SIZE) - kfree(hash); + kfree(buckets); else - free_pages((unsigned long)hash, get_order(size)); + free_pages((unsigned long)buckets, get_order(size)); + kfree(nht); } -static void neigh_hash_grow(struct neigh_table *tbl, unsigned long new_entries) +static struct neigh_hash_table *neigh_hash_grow(struct neigh_table *tbl, + unsigned long new_entries) { - struct neighbour **new_hash, **old_hash; - unsigned int i, new_hash_mask, old_entries; + unsigned int i, hash; + struct neigh_hash_table *new_nht, *old_nht; NEIGH_CACHE_STAT_INC(tbl, hash_grows); BUG_ON(!is_power_of_2(new_entries)); - new_hash = neigh_hash_alloc(new_entries); - if (!new_hash) - return; + old_nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); + new_nht = neigh_hash_alloc(new_entries); + if (!new_nht) + return old_nht; - old_entries = tbl->hash_mask + 1; - new_hash_mask = new_entries - 1; - old_hash = tbl->hash_buckets; - - get_random_bytes(&tbl->hash_rnd, sizeof(tbl->hash_rnd)); - for (i = 0; i < old_entries; i++) { + for (i = 0; i <= old_nht->hash_mask; i++) { struct neighbour *n, *next; - for (n = old_hash[i]; n; n = next) { - unsigned int hash_val = tbl->hash(n->primary_key, n->dev); + for (n = old_nht->hash_buckets[i]; + n != NULL; + n = next) { + hash = tbl->hash(n->primary_key, n->dev, + new_nht->hash_rnd); - hash_val &= new_hash_mask; + hash &= new_nht->hash_mask; next = n->next; - n->next = new_hash[hash_val]; - new_hash[hash_val] = n; + n->next = new_nht->hash_buckets[hash]; + new_nht->hash_buckets[hash] = n; } } - tbl->hash_buckets = new_hash; - tbl->hash_mask = new_hash_mask; - neigh_hash_free(old_hash, old_entries); + rcu_assign_pointer(tbl->nht, new_nht); + call_rcu(&old_nht->rcu, neigh_hash_free_rcu); + return new_nht; } struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey, @@ -363,19 +387,23 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey, struct neighbour *n; int key_len = tbl->key_len; u32 hash_val; + struct neigh_hash_table *nht; NEIGH_CACHE_STAT_INC(tbl, lookups); - read_lock_bh(&tbl->lock); - hash_val = tbl->hash(pkey, dev); - for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) { + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + hash_val = tbl->hash(pkey, dev, nht->hash_rnd) & nht->hash_mask; + read_lock(&tbl->lock); + for (n = nht->hash_buckets[hash_val]; n; n = n->next) { if (dev == n->dev && !memcmp(n->primary_key, pkey, key_len)) { neigh_hold(n); NEIGH_CACHE_STAT_INC(tbl, hits); break; } } - read_unlock_bh(&tbl->lock); + read_unlock(&tbl->lock); + rcu_read_unlock_bh(); return n; } EXPORT_SYMBOL(neigh_lookup); @@ -386,12 +414,15 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net, struct neighbour *n; int key_len = tbl->key_len; u32 hash_val; + struct neigh_hash_table *nht; NEIGH_CACHE_STAT_INC(tbl, lookups); - read_lock_bh(&tbl->lock); - hash_val = tbl->hash(pkey, NULL); - for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) { + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + hash_val = tbl->hash(pkey, NULL, nht->hash_rnd) & nht->hash_mask; + read_lock(&tbl->lock); + for (n = nht->hash_buckets[hash_val]; n; n = n->next) { if (!memcmp(n->primary_key, pkey, key_len) && net_eq(dev_net(n->dev), net)) { neigh_hold(n); @@ -399,7 +430,8 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net, break; } } - read_unlock_bh(&tbl->lock); + read_unlock(&tbl->lock); + rcu_read_unlock_bh(); return n; } EXPORT_SYMBOL(neigh_lookup_nodev); @@ -411,6 +443,7 @@ struct neighbour *neigh_create(struct neigh_table *tbl, const void *pkey, int key_len = tbl->key_len; int error; struct neighbour *n1, *rc, *n = neigh_alloc(tbl); + struct neigh_hash_table *nht; if (!n) { rc = ERR_PTR(-ENOBUFS); @@ -437,18 +470,20 @@ struct neighbour *neigh_create(struct neigh_table *tbl, const void *pkey, n->confirmed = jiffies - (n->parms->base_reachable_time << 1); write_lock_bh(&tbl->lock); + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); - if (atomic_read(&tbl->entries) > (tbl->hash_mask + 1)) - neigh_hash_grow(tbl, (tbl->hash_mask + 1) << 1); + if (atomic_read(&tbl->entries) > (nht->hash_mask + 1)) + nht = neigh_hash_grow(tbl, (nht->hash_mask + 1) << 1); - hash_val = tbl->hash(pkey, dev) & tbl->hash_mask; + hash_val = tbl->hash(pkey, dev, nht->hash_rnd) & nht->hash_mask; if (n->parms->dead) { rc = ERR_PTR(-EINVAL); goto out_tbl_unlock; } - for (n1 = tbl->hash_buckets[hash_val]; n1; n1 = n1->next) { + for (n1 = nht->hash_buckets[hash_val]; n1; n1 = n1->next) { if (dev == n1->dev && !memcmp(n1->primary_key, pkey, key_len)) { neigh_hold(n1); rc = n1; @@ -456,8 +491,8 @@ struct neighbour *neigh_create(struct neigh_table *tbl, const void *pkey, } } - n->next = tbl->hash_buckets[hash_val]; - tbl->hash_buckets[hash_val] = n; + n->next = nht->hash_buckets[hash_val]; + nht->hash_buckets[hash_val] = n; n->dead = 0; neigh_hold(n); write_unlock_bh(&tbl->lock); @@ -698,10 +733,13 @@ static void neigh_periodic_work(struct work_struct *work) struct neigh_table *tbl = container_of(work, struct neigh_table, gc_work.work); struct neighbour *n, **np; unsigned int i; + struct neigh_hash_table *nht; NEIGH_CACHE_STAT_INC(tbl, periodic_gc_runs); write_lock_bh(&tbl->lock); + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); /* * periodically recompute ReachableTime from random function @@ -715,8 +753,8 @@ static void neigh_periodic_work(struct work_struct *work) neigh_rand_reach_time(p->base_reachable_time); } - for (i = 0 ; i <= tbl->hash_mask; i++) { - np = &tbl->hash_buckets[i]; + for (i = 0 ; i <= nht->hash_mask; i++) { + np = &nht->hash_buckets[i]; while ((n = *np) != NULL) { unsigned int state; @@ -1438,17 +1476,14 @@ void neigh_table_init_no_netlink(struct neigh_table *tbl) panic("cannot create neighbour proc dir entry"); #endif - tbl->hash_mask = 1; - tbl->hash_buckets = neigh_hash_alloc(tbl->hash_mask + 1); + tbl->nht = neigh_hash_alloc(8); phsize = (PNEIGH_HASHMASK + 1) * sizeof(struct pneigh_entry *); tbl->phash_buckets = kzalloc(phsize, GFP_KERNEL); - if (!tbl->hash_buckets || !tbl->phash_buckets) + if (!tbl->nht || !tbl->phash_buckets) panic("cannot allocate neighbour cache hashes"); - get_random_bytes(&tbl->hash_rnd, sizeof(tbl->hash_rnd)); - rwlock_init(&tbl->lock); INIT_DELAYED_WORK_DEFERRABLE(&tbl->gc_work, neigh_periodic_work); schedule_delayed_work(&tbl->gc_work, tbl->parms.reachable_time); @@ -1504,8 +1539,8 @@ int neigh_table_clear(struct neigh_table *tbl) } write_unlock(&neigh_tbl_lock); - neigh_hash_free(tbl->hash_buckets, tbl->hash_mask + 1); - tbl->hash_buckets = NULL; + call_rcu(&tbl->nht->rcu, neigh_hash_free_rcu); + tbl->nht = NULL; kfree(tbl->phash_buckets); tbl->phash_buckets = NULL; @@ -1745,18 +1780,22 @@ static int neightbl_fill_info(struct sk_buff *skb, struct neigh_table *tbl, unsigned long now = jiffies; unsigned int flush_delta = now - tbl->last_flush; unsigned int rand_delta = now - tbl->last_rand; - + struct neigh_hash_table *nht; struct ndt_config ndc = { .ndtc_key_len = tbl->key_len, .ndtc_entry_size = tbl->entry_size, .ndtc_entries = atomic_read(&tbl->entries), .ndtc_last_flush = jiffies_to_msecs(flush_delta), .ndtc_last_rand = jiffies_to_msecs(rand_delta), - .ndtc_hash_rnd = tbl->hash_rnd, - .ndtc_hash_mask = tbl->hash_mask, .ndtc_proxy_qlen = tbl->proxy_queue.qlen, }; + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + ndc.ndtc_hash_rnd = nht->hash_rnd; + ndc.ndtc_hash_mask = nht->hash_mask; + rcu_read_unlock_bh(); + NLA_PUT(skb, NDTA_CONFIG, sizeof(ndc), &ndc); } @@ -2088,14 +2127,18 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, struct neighbour *n; int rc, h, s_h = cb->args[1]; int idx, s_idx = idx = cb->args[2]; + struct neigh_hash_table *nht; - read_lock_bh(&tbl->lock); - for (h = 0; h <= tbl->hash_mask; h++) { + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + + read_lock(&tbl->lock); + for (h = 0; h <= nht->hash_mask; h++) { if (h < s_h) continue; if (h > s_h) s_idx = 0; - for (n = tbl->hash_buckets[h], idx = 0; n; n = n->next) { + for (n = nht->hash_buckets[h], idx = 0; n; n = n->next) { if (!net_eq(dev_net(n->dev), net)) continue; if (idx < s_idx) @@ -2104,7 +2147,6 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, cb->nlh->nlmsg_seq, RTM_NEWNEIGH, NLM_F_MULTI) <= 0) { - read_unlock_bh(&tbl->lock); rc = -1; goto out; } @@ -2112,9 +2154,10 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, idx++; } } - read_unlock_bh(&tbl->lock); rc = skb->len; out: + read_unlock(&tbl->lock); + rcu_read_unlock_bh(); cb->args[1] = h; cb->args[2] = idx; return rc; @@ -2147,15 +2190,20 @@ static int neigh_dump_info(struct sk_buff *skb, struct netlink_callback *cb) void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void *), void *cookie) { int chain; + struct neigh_hash_table *nht; - read_lock_bh(&tbl->lock); - for (chain = 0; chain <= tbl->hash_mask; chain++) { + rcu_read_lock_bh(); + nht = rcu_dereference_bh(tbl->nht); + + read_lock(&tbl->lock); + for (chain = 0; chain <= nht->hash_mask; chain++) { struct neighbour *n; - for (n = tbl->hash_buckets[chain]; n; n = n->next) + for (n = nht->hash_buckets[chain]; n; n = n->next) cb(n, cookie); } - read_unlock_bh(&tbl->lock); + read_unlock(&tbl->lock); + rcu_read_unlock_bh(); } EXPORT_SYMBOL(neigh_for_each); @@ -2164,11 +2212,14 @@ void __neigh_for_each_release(struct neigh_table *tbl, int (*cb)(struct neighbour *)) { int chain; + struct neigh_hash_table *nht; - for (chain = 0; chain <= tbl->hash_mask; chain++) { + nht = rcu_dereference_protected(tbl->nht, + lockdep_is_held(&tbl->lock)); + for (chain = 0; chain <= nht->hash_mask; chain++) { struct neighbour *n, **np; - np = &tbl->hash_buckets[chain]; + np = &nht->hash_buckets[chain]; while ((n = *np) != NULL) { int release; @@ -2193,13 +2244,13 @@ static struct neighbour *neigh_get_first(struct seq_file *seq) { struct neigh_seq_state *state = seq->private; struct net *net = seq_file_net(seq); - struct neigh_table *tbl = state->tbl; + struct neigh_hash_table *nht = state->nht; struct neighbour *n = NULL; int bucket = state->bucket; state->flags &= ~NEIGH_SEQ_IS_PNEIGH; - for (bucket = 0; bucket <= tbl->hash_mask; bucket++) { - n = tbl->hash_buckets[bucket]; + for (bucket = 0; bucket <= nht->hash_mask; bucket++) { + n = nht->hash_buckets[bucket]; while (n) { if (!net_eq(dev_net(n->dev), net)) @@ -2234,7 +2285,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, { struct neigh_seq_state *state = seq->private; struct net *net = seq_file_net(seq); - struct neigh_table *tbl = state->tbl; + struct neigh_hash_table *nht = state->nht; if (state->neigh_sub_iter) { void *v = state->neigh_sub_iter(state, n, pos); @@ -2265,10 +2316,10 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, if (n) break; - if (++state->bucket > tbl->hash_mask) + if (++state->bucket > nht->hash_mask) break; - n = tbl->hash_buckets[state->bucket]; + n = nht->hash_buckets[state->bucket]; } if (n && pos) @@ -2367,6 +2418,7 @@ static void *neigh_get_idx_any(struct seq_file *seq, loff_t *pos) void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl, unsigned int neigh_seq_flags) __acquires(tbl->lock) + __acquires(rcu_bh) { struct neigh_seq_state *state = seq->private; @@ -2374,8 +2426,9 @@ void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl state->bucket = 0; state->flags = (neigh_seq_flags & ~NEIGH_SEQ_IS_PNEIGH); - read_lock_bh(&tbl->lock); - + rcu_read_lock_bh(); + state->nht = rcu_dereference_bh(tbl->nht); + read_lock(&tbl->lock); return *pos ? neigh_get_idx_any(seq, pos) : SEQ_START_TOKEN; } EXPORT_SYMBOL(neigh_seq_start); @@ -2409,11 +2462,13 @@ EXPORT_SYMBOL(neigh_seq_next); void neigh_seq_stop(struct seq_file *seq, void *v) __releases(tbl->lock) + __releases(rcu_bh) { struct neigh_seq_state *state = seq->private; struct neigh_table *tbl = state->tbl; - read_unlock_bh(&tbl->lock); + read_unlock(&tbl->lock); + rcu_read_unlock_bh(); } EXPORT_SYMBOL(neigh_seq_stop); diff --git a/net/decnet/dn_neigh.c b/net/decnet/dn_neigh.c index 0363bb95cc7d..a085dbcf5c7f 100644 --- a/net/decnet/dn_neigh.c +++ b/net/decnet/dn_neigh.c @@ -48,7 +48,6 @@ #include #include -static u32 dn_neigh_hash(const void *pkey, const struct net_device *dev); static int dn_neigh_construct(struct neighbour *); static void dn_long_error_report(struct neighbour *, struct sk_buff *); static void dn_short_error_report(struct neighbour *, struct sk_buff *); @@ -93,6 +92,13 @@ static const struct neigh_ops dn_phase3_ops = { .queue_xmit = dev_queue_xmit }; +static u32 dn_neigh_hash(const void *pkey, + const struct net_device *dev, + __u32 hash_rnd) +{ + return jhash_2words(*(__u16 *)pkey, 0, hash_rnd); +} + struct neigh_table dn_neigh_table = { .family = PF_DECnet, .entry_size = sizeof(struct dn_neigh), @@ -122,11 +128,6 @@ struct neigh_table dn_neigh_table = { .gc_thresh3 = 1024, }; -static u32 dn_neigh_hash(const void *pkey, const struct net_device *dev) -{ - return jhash_2words(*(__u16 *)pkey, 0, dn_neigh_table.hash_rnd); -} - static int dn_neigh_construct(struct neighbour *neigh) { struct net_device *dev = neigh->dev; diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index d9031ad67826..f35309578170 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -127,7 +127,7 @@ EXPORT_SYMBOL(clip_tbl_hook); /* * Interface to generic neighbour cache. */ -static u32 arp_hash(const void *pkey, const struct net_device *dev); +static u32 arp_hash(const void *pkey, const struct net_device *dev, __u32 rnd); static int arp_constructor(struct neighbour *neigh); static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb); static void arp_error_report(struct neighbour *neigh, struct sk_buff *skb); @@ -225,9 +225,11 @@ int arp_mc_map(__be32 addr, u8 *haddr, struct net_device *dev, int dir) } -static u32 arp_hash(const void *pkey, const struct net_device *dev) +static u32 arp_hash(const void *pkey, + const struct net_device *dev, + __u32 hash_rnd) { - return jhash_2words(*(u32 *)pkey, dev->ifindex, arp_tbl.hash_rnd); + return jhash_2words(*(u32 *)pkey, dev->ifindex, hash_rnd); } static int arp_constructor(struct neighbour *neigh) diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index b3dd844cd34f..998d6d27e7cf 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -91,7 +91,9 @@ #include #include -static u32 ndisc_hash(const void *pkey, const struct net_device *dev); +static u32 ndisc_hash(const void *pkey, + const struct net_device *dev, + __u32 rnd); static int ndisc_constructor(struct neighbour *neigh); static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb); static void ndisc_error_report(struct neighbour *neigh, struct sk_buff *skb); @@ -350,7 +352,9 @@ int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int d EXPORT_SYMBOL(ndisc_mc_map); -static u32 ndisc_hash(const void *pkey, const struct net_device *dev) +static u32 ndisc_hash(const void *pkey, + const struct net_device *dev, + __u32 hash_rnd) { const u32 *p32 = pkey; u32 addr_hash, i; @@ -359,7 +363,7 @@ static u32 ndisc_hash(const void *pkey, const struct net_device *dev) for (i = 0; i < (sizeof(struct in6_addr) / sizeof(u32)); i++) addr_hash ^= *p32++; - return jhash_2words(addr_hash, dev->ifindex, nd_tbl.hash_rnd); + return jhash_2words(addr_hash, dev->ifindex, hash_rnd); } static int ndisc_constructor(struct neighbour *neigh) -- cgit v1.2.3 From ebc0ffae5dfb4447e0a431ffe7fe1d467c48bbb9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 5 Oct 2010 10:41:36 +0000 Subject: fib: RCU conversion of fib_lookup() fib_lookup() converted to be called in RCU protected context, no reference taken and released on a contended cache line (fib_clntref) fib_table_lookup() and fib_semantic_match() get an additional parameter. struct fib_info gets an rcu_head field, and is freed after an rcu grace period. Stress test : (Sending 160.000.000 UDP frames on same neighbour, IP route cache disabled, dual E5540 @2.53GHz, 32bit kernel, FIB_HASH) (about same results for FIB_TRIE) Before patch : real 1m31.199s user 0m13.761s sys 23m24.780s After patch: real 1m5.375s user 0m14.997s sys 15m50.115s Before patch Profile : 13044.00 15.4% __ip_route_output_key vmlinux 8438.00 10.0% dst_destroy vmlinux 5983.00 7.1% fib_semantic_match vmlinux 5410.00 6.4% fib_rules_lookup vmlinux 4803.00 5.7% neigh_lookup vmlinux 4420.00 5.2% _raw_spin_lock vmlinux 3883.00 4.6% rt_set_nexthop vmlinux 3261.00 3.9% _raw_read_lock vmlinux 2794.00 3.3% fib_table_lookup vmlinux 2374.00 2.8% neigh_resolve_output vmlinux 2153.00 2.5% dst_alloc vmlinux 1502.00 1.8% _raw_read_lock_bh vmlinux 1484.00 1.8% kmem_cache_alloc vmlinux 1407.00 1.7% eth_header vmlinux 1406.00 1.7% ipv4_dst_destroy vmlinux 1298.00 1.5% __copy_from_user_ll vmlinux 1174.00 1.4% dev_queue_xmit vmlinux 1000.00 1.2% ip_output vmlinux After patch Profile : 13712.00 15.8% dst_destroy vmlinux 8548.00 9.9% __ip_route_output_key vmlinux 7017.00 8.1% neigh_lookup vmlinux 4554.00 5.3% fib_semantic_match vmlinux 4067.00 4.7% _raw_read_lock vmlinux 3491.00 4.0% dst_alloc vmlinux 3186.00 3.7% neigh_resolve_output vmlinux 3103.00 3.6% fib_table_lookup vmlinux 2098.00 2.4% _raw_read_lock_bh vmlinux 2081.00 2.4% kmem_cache_alloc vmlinux 2013.00 2.3% _raw_spin_lock vmlinux 1763.00 2.0% __copy_from_user_ll vmlinux 1763.00 2.0% ip_output vmlinux 1761.00 2.0% ipv4_dst_destroy vmlinux 1631.00 1.9% eth_header vmlinux 1440.00 1.7% _raw_read_unlock_bh vmlinux Reference results, if IP route cache is enabled : real 0m29.718s user 0m10.845s sys 7m37.341s 25213.00 29.5% __ip_route_output_key vmlinux 9011.00 10.5% dst_release vmlinux 4817.00 5.6% ip_push_pending_frames vmlinux 4232.00 5.0% ip_finish_output vmlinux 3940.00 4.6% udp_sendmsg vmlinux 3730.00 4.4% __copy_from_user_ll vmlinux 3716.00 4.4% ip_route_output_flow vmlinux 2451.00 2.9% __xfrm_lookup vmlinux 2221.00 2.6% ip_append_data vmlinux 1718.00 2.0% _raw_spin_lock_bh vmlinux 1655.00 1.9% __alloc_skb vmlinux 1572.00 1.8% sock_wfree vmlinux 1345.00 1.6% kfree vmlinux Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/fib_rules.h | 2 ++ include/net/ip_fib.h | 17 ++++---------- net/core/fib_rules.c | 3 ++- net/ipv4/fib_frontend.c | 27 +++++++++++----------- net/ipv4/fib_hash.c | 5 ++-- net/ipv4/fib_lookup.h | 2 +- net/ipv4/fib_rules.c | 3 ++- net/ipv4/fib_semantics.c | 21 +++++++++++++---- net/ipv4/fib_trie.c | 10 ++++---- net/ipv4/route.c | 59 ++++++++++++++++++++---------------------------- 10 files changed, 72 insertions(+), 77 deletions(-) (limited to 'net/core') diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index ac2fd002812e..106f3097d384 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -31,6 +31,8 @@ struct fib_lookup_arg { void *lookup_ptr; void *result; struct fib_rule *rule; + int flags; +#define FIB_LOOKUP_NOREF 1 }; struct fib_rules_ops { diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index c93f94edc610..ba3666d31766 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -86,6 +86,7 @@ struct fib_info { #ifdef CONFIG_IP_ROUTE_MULTIPATH int fib_power; #endif + struct rcu_head rcu; struct fib_nh fib_nh[0]; #define fib_dev fib_nh[0].nh_dev }; @@ -148,7 +149,7 @@ struct fib_table { }; extern int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, - struct fib_result *res); + struct fib_result *res, int fib_flags); extern int fib_table_insert(struct fib_table *, struct fib_config *); extern int fib_table_delete(struct fib_table *, struct fib_config *); extern int fib_table_dump(struct fib_table *table, struct sk_buff *skb, @@ -185,11 +186,11 @@ static inline int fib_lookup(struct net *net, const struct flowi *flp, struct fib_table *table; table = fib_get_table(net, RT_TABLE_LOCAL); - if (!fib_table_lookup(table, flp, res)) + if (!fib_table_lookup(table, flp, res, FIB_LOOKUP_NOREF)) return 0; table = fib_get_table(net, RT_TABLE_MAIN); - if (!fib_table_lookup(table, flp, res)) + if (!fib_table_lookup(table, flp, res, FIB_LOOKUP_NOREF)) return 0; return -ENETUNREACH; } @@ -254,16 +255,6 @@ static inline void fib_info_put(struct fib_info *fi) free_fib_info(fi); } -static inline void fib_res_put(struct fib_result *res) -{ - if (res->fi) - fib_info_put(res->fi); -#ifdef CONFIG_IP_MULTIPLE_TABLES - if (res->r) - fib_rule_put(res->r); -#endif -} - #ifdef CONFIG_PROC_FS extern int __net_init fib_proc_init(struct net *net); extern void __net_exit fib_proc_exit(struct net *net); diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index cfb7d25c172d..21698f8c49ee 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -225,7 +225,8 @@ jumped: err = ops->action(rule, fl, flags, arg); if (err != -EAGAIN) { - if (likely(atomic_inc_not_zero(&rule->refcnt))) { + if ((arg->flags & FIB_LOOKUP_NOREF) || + likely(atomic_inc_not_zero(&rule->refcnt))) { arg->rule = rule; goto out; } diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index b05c23b05a9f..919f2ad19b49 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -168,8 +168,11 @@ struct net_device *__ip_dev_find(struct net *net, __be32 addr, bool devref) struct fib_result res = { 0 }; struct net_device *dev = NULL; - if (fib_lookup(net, &fl, &res)) + rcu_read_lock(); + if (fib_lookup(net, &fl, &res)) { + rcu_read_unlock(); return NULL; + } if (res.type != RTN_LOCAL) goto out; dev = FIB_RES_DEV(res); @@ -177,7 +180,7 @@ struct net_device *__ip_dev_find(struct net *net, __be32 addr, bool devref) if (dev && devref) dev_hold(dev); out: - fib_res_put(&res); + rcu_read_unlock(); return dev; } EXPORT_SYMBOL(__ip_dev_find); @@ -207,11 +210,12 @@ static inline unsigned __inet_dev_addr_type(struct net *net, local_table = fib_get_table(net, RT_TABLE_LOCAL); if (local_table) { ret = RTN_UNICAST; - if (!fib_table_lookup(local_table, &fl, &res)) { + rcu_read_lock(); + if (!fib_table_lookup(local_table, &fl, &res, FIB_LOOKUP_NOREF)) { if (!dev || dev == res.fi->fib_dev) ret = res.type; - fib_res_put(&res); } + rcu_read_unlock(); } return ret; } @@ -235,6 +239,7 @@ EXPORT_SYMBOL(inet_dev_addr_type); * - figure out what "logical" interface this packet arrived * and calculate "specific destination" address. * - check, that packet arrived from expected physical interface. + * called with rcu_read_lock() */ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, struct net_device *dev, __be32 *spec_dst, @@ -259,7 +264,6 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, struct net *net; no_addr = rpf = accept_local = 0; - rcu_read_lock(); in_dev = __in_dev_get_rcu(dev); if (in_dev) { no_addr = in_dev->ifa_list == NULL; @@ -268,7 +272,6 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, if (mark && !IN_DEV_SRC_VMARK(in_dev)) fl.mark = 0; } - rcu_read_unlock(); if (in_dev == NULL) goto e_inval; @@ -278,7 +281,7 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, goto last_resort; if (res.type != RTN_UNICAST) { if (res.type != RTN_LOCAL || !accept_local) - goto e_inval_res; + goto e_inval; } *spec_dst = FIB_RES_PREFSRC(res); fib_combine_itag(itag, &res); @@ -299,10 +302,8 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, #endif if (dev_match) { ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST; - fib_res_put(&res); return ret; } - fib_res_put(&res); if (no_addr) goto last_resort; if (rpf == 1) @@ -315,7 +316,6 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, *spec_dst = FIB_RES_PREFSRC(res); ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST; } - fib_res_put(&res); } return ret; @@ -326,8 +326,6 @@ last_resort: *itag = 0; return 0; -e_inval_res: - fib_res_put(&res); e_inval: return -EINVAL; e_rpf: @@ -873,15 +871,16 @@ static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb) local_bh_disable(); frn->tb_id = tb->tb_id; - frn->err = fib_table_lookup(tb, &fl, &res); + rcu_read_lock(); + frn->err = fib_table_lookup(tb, &fl, &res, FIB_LOOKUP_NOREF); if (!frn->err) { frn->prefixlen = res.prefixlen; frn->nh_sel = res.nh_sel; frn->type = res.type; frn->scope = res.scope; - fib_res_put(&res); } + rcu_read_unlock(); local_bh_enable(); } } diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c index 4ed7e0dea1bc..83cca68e259c 100644 --- a/net/ipv4/fib_hash.c +++ b/net/ipv4/fib_hash.c @@ -244,7 +244,8 @@ fn_new_zone(struct fn_hash *table, int z) } int fib_table_lookup(struct fib_table *tb, - const struct flowi *flp, struct fib_result *res) + const struct flowi *flp, struct fib_result *res, + int fib_flags) { int err; struct fn_zone *fz; @@ -264,7 +265,7 @@ int fib_table_lookup(struct fib_table *tb, err = fib_semantic_match(&f->fn_alias, flp, res, - fz->fz_order); + fz->fz_order, fib_flags); if (err <= 0) goto out; } diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h index 637b133973bd..b9c9a9f2aee5 100644 --- a/net/ipv4/fib_lookup.h +++ b/net/ipv4/fib_lookup.h @@ -22,7 +22,7 @@ struct fib_alias { /* Exported by fib_semantics.c */ extern int fib_semantic_match(struct list_head *head, const struct flowi *flp, - struct fib_result *res, int prefixlen); + struct fib_result *res, int prefixlen, int fib_flags); extern void fib_release_info(struct fib_info *); extern struct fib_info *fib_create_info(struct fib_config *cfg); extern int fib_nh_match(struct fib_config *cfg, struct fib_info *fi); diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index 32300521e32c..7981a24f5c7b 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -57,6 +57,7 @@ int fib_lookup(struct net *net, struct flowi *flp, struct fib_result *res) { struct fib_lookup_arg arg = { .result = res, + .flags = FIB_LOOKUP_NOREF, }; int err; @@ -94,7 +95,7 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, if (!tbl) goto errout; - err = fib_table_lookup(tbl, flp, (struct fib_result *) arg->result); + err = fib_table_lookup(tbl, flp, (struct fib_result *) arg->result, arg->flags); if (err > 0) err = -EAGAIN; errout: diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index ba52f399a898..0f80dfc2f7fb 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -148,6 +148,13 @@ static const struct /* Release a nexthop info record */ +static void free_fib_info_rcu(struct rcu_head *head) +{ + struct fib_info *fi = container_of(head, struct fib_info, rcu); + + kfree(fi); +} + void free_fib_info(struct fib_info *fi) { if (fi->fib_dead == 0) { @@ -161,7 +168,7 @@ void free_fib_info(struct fib_info *fi) } endfor_nexthops(fi); fib_info_cnt--; release_net(fi->fib_net); - kfree(fi); + call_rcu(&fi->rcu, free_fib_info_rcu); } void fib_release_info(struct fib_info *fi) @@ -553,6 +560,7 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi, nh->nh_scope = RT_SCOPE_LINK; return 0; } + rcu_read_lock(); { struct flowi fl = { .nl_u = { @@ -568,8 +576,10 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi, if (fl.fl4_scope < RT_SCOPE_LINK) fl.fl4_scope = RT_SCOPE_LINK; err = fib_lookup(net, &fl, &res); - if (err) + if (err) { + rcu_read_unlock(); return err; + } } err = -EINVAL; if (res.type != RTN_UNICAST && res.type != RTN_LOCAL) @@ -585,7 +595,7 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi, goto out; err = 0; out: - fib_res_put(&res); + rcu_read_unlock(); return err; } else { struct in_device *in_dev; @@ -879,7 +889,7 @@ failure: /* Note! fib_semantic_match intentionally uses RCU list functions. */ int fib_semantic_match(struct list_head *head, const struct flowi *flp, - struct fib_result *res, int prefixlen) + struct fib_result *res, int prefixlen, int fib_flags) { struct fib_alias *fa; int nh_sel = 0; @@ -943,7 +953,8 @@ out_fill_res: res->type = fa->fa_type; res->scope = fa->fa_scope; res->fi = fa->fa_info; - atomic_inc(&res->fi->fib_clntref); + if (!(fib_flags & FIB_LOOKUP_NOREF)) + atomic_inc(&res->fi->fib_clntref); return 0; } diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index a96e5ec211a0..271c89bdf049 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1342,7 +1342,7 @@ err: /* should be called with rcu_read_lock */ static int check_leaf(struct trie *t, struct leaf *l, t_key key, const struct flowi *flp, - struct fib_result *res) + struct fib_result *res, int fib_flags) { struct leaf_info *li; struct hlist_head *hhead = &l->list; @@ -1356,7 +1356,7 @@ static int check_leaf(struct trie *t, struct leaf *l, if (l->key != (key & ntohl(mask))) continue; - err = fib_semantic_match(&li->falh, flp, res, plen); + err = fib_semantic_match(&li->falh, flp, res, plen, fib_flags); #ifdef CONFIG_IP_FIB_TRIE_STATS if (err <= 0) @@ -1372,7 +1372,7 @@ static int check_leaf(struct trie *t, struct leaf *l, } int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, - struct fib_result *res) + struct fib_result *res, int fib_flags) { struct trie *t = (struct trie *) tb->tb_data; int ret; @@ -1399,7 +1399,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, /* Just a leaf? */ if (IS_LEAF(n)) { - ret = check_leaf(t, (struct leaf *)n, key, flp, res); + ret = check_leaf(t, (struct leaf *)n, key, flp, res, fib_flags); goto found; } @@ -1424,7 +1424,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi *flp, } if (IS_LEAF(n)) { - ret = check_leaf(t, (struct leaf *)n, key, flp, res); + ret = check_leaf(t, (struct leaf *)n, key, flp, res, fib_flags); if (ret > 0) goto backtrace; goto found; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 04e0df82b88c..7864d0c48968 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1773,12 +1773,15 @@ void ip_rt_get_source(u8 *addr, struct rtable *rt) if (rt->fl.iif == 0) src = rt->rt_src; - else if (fib_lookup(dev_net(rt->dst.dev), &rt->fl, &res) == 0) { - src = FIB_RES_PREFSRC(res); - fib_res_put(&res); - } else - src = inet_select_addr(rt->dst.dev, rt->rt_gateway, + else { + rcu_read_lock(); + if (fib_lookup(dev_net(rt->dst.dev), &rt->fl, &res) == 0) + src = FIB_RES_PREFSRC(res); + else + src = inet_select_addr(rt->dst.dev, rt->rt_gateway, RT_SCOPE_UNIVERSE); + rcu_read_unlock(); + } memcpy(addr, &src, 4); } @@ -2081,6 +2084,7 @@ static int ip_mkroute_input(struct sk_buff *skb, * Such approach solves two big problems: * 1. Not simplex devices are handled properly. * 2. IP spoofing attempts are filtered with 100% of guarantee. + * called with rcu_read_lock() */ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, @@ -2102,7 +2106,6 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, unsigned hash; __be32 spec_dst; int err = -EINVAL; - int free_res = 0; struct net * net = dev_net(dev); /* IP on this device is disabled. */ @@ -2134,12 +2137,12 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, /* * Now we are ready to route packet. */ - if ((err = fib_lookup(net, &fl, &res)) != 0) { + err = fib_lookup(net, &fl, &res); + if (err != 0) { if (!IN_DEV_FORWARD(in_dev)) goto e_hostunreach; goto no_route; } - free_res = 1; RT_CACHE_STAT_INC(in_slow_tot); @@ -2148,8 +2151,8 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, if (res.type == RTN_LOCAL) { err = fib_validate_source(saddr, daddr, tos, - net->loopback_dev->ifindex, - dev, &spec_dst, &itag, skb->mark); + net->loopback_dev->ifindex, + dev, &spec_dst, &itag, skb->mark); if (err < 0) goto martian_source_keep_err; if (err) @@ -2164,9 +2167,6 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, goto martian_destination; err = ip_mkroute_input(skb, &res, &fl, in_dev, daddr, saddr, tos); -done: - if (free_res) - fib_res_put(&res); out: return err; brd_input: @@ -2226,7 +2226,7 @@ local_input: rth->rt_type = res.type; hash = rt_hash(daddr, saddr, fl.iif, rt_genid(net)); err = rt_intern_hash(hash, rth, NULL, skb, fl.iif); - goto done; + goto out; no_route: RT_CACHE_STAT_INC(in_no_route); @@ -2249,21 +2249,21 @@ martian_destination: e_hostunreach: err = -EHOSTUNREACH; - goto done; + goto out; e_inval: err = -EINVAL; - goto done; + goto out; e_nobufs: err = -ENOBUFS; - goto done; + goto out; martian_source: err = -EINVAL; martian_source_keep_err: ip_handle_martian_source(dev, in_dev, skb, daddr, saddr); - goto done; + goto out; } int ip_route_input_common(struct sk_buff *skb, __be32 daddr, __be32 saddr, @@ -2349,6 +2349,7 @@ skip_cache: } EXPORT_SYMBOL(ip_route_input_common); +/* called with rcu_read_lock() */ static int __mkroute_output(struct rtable **result, struct fib_result *res, const struct flowi *fl, @@ -2373,18 +2374,13 @@ static int __mkroute_output(struct rtable **result, if (dev_out->flags & IFF_LOOPBACK) flags |= RTCF_LOCAL; - rcu_read_lock(); in_dev = __in_dev_get_rcu(dev_out); - if (!in_dev) { - rcu_read_unlock(); + if (!in_dev) return -EINVAL; - } + if (res->type == RTN_BROADCAST) { flags |= RTCF_BROADCAST | RTCF_LOCAL; - if (res->fi) { - fib_info_put(res->fi); - res->fi = NULL; - } + res->fi = NULL; } else if (res->type == RTN_MULTICAST) { flags |= RTCF_MULTICAST | RTCF_LOCAL; if (!ip_check_mc(in_dev, oldflp->fl4_dst, oldflp->fl4_src, @@ -2394,10 +2390,8 @@ static int __mkroute_output(struct rtable **result, * default one, but do not gateway in this case. * Yes, it is hack. */ - if (res->fi && res->prefixlen < 4) { - fib_info_put(res->fi); + if (res->fi && res->prefixlen < 4) res->fi = NULL; - } } @@ -2467,6 +2461,7 @@ static int __mkroute_output(struct rtable **result, return 0; } +/* called with rcu_read_lock() */ static int ip_mkroute_output(struct rtable **rp, struct fib_result *res, const struct flowi *fl, @@ -2509,7 +2504,6 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp, struct fib_result res; unsigned int flags = 0; struct net_device *dev_out = NULL; - int free_res = 0; int err; @@ -2636,15 +2630,12 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp, err = -ENETUNREACH; goto out; } - free_res = 1; if (res.type == RTN_LOCAL) { if (!fl.fl4_src) fl.fl4_src = fl.fl4_dst; dev_out = net->loopback_dev; fl.oif = dev_out->ifindex; - if (res.fi) - fib_info_put(res.fi); res.fi = NULL; flags |= RTCF_LOCAL; goto make_route; @@ -2668,8 +2659,6 @@ static int ip_route_output_slow(struct net *net, struct rtable **rp, make_route: err = ip_mkroute_output(rp, &res, &fl, oldflp, dev_out, flags); - if (free_res) - fib_res_put(&res); out: return err; } -- cgit v1.2.3 From 767e97e1e0db0d0f3152cd2f3bd3403596aedbad Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 6 Oct 2010 17:49:21 -0700 Subject: neigh: RCU conversion of struct neighbour This is the second step for neighbour RCU conversion. (first was commit d6bf7817 : RCU conversion of neigh hash table) neigh_lookup() becomes lockless, but still take a reference on found neighbour. (no more read_lock()/read_unlock() on tbl->lock) struct neighbour gets an additional rcu_head field and is freed after an RCU grace period. Future work would need to eventually not take a reference on neighbour for temporary dst (DST_NOCACHE), but this would need dst->_neighbour to use a noref bit like we did for skb->_dst. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/neighbour.h | 5 +- net/core/neighbour.c | 137 ++++++++++++++++++++++++++++++------------------ 2 files changed, 88 insertions(+), 54 deletions(-) (limited to 'net/core') diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 37845dae6488..a4538d553704 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -91,7 +91,7 @@ struct neigh_statistics { #define NEIGH_CACHE_STAT_INC(tbl, field) this_cpu_inc((tbl)->stats->field) struct neighbour { - struct neighbour *next; + struct neighbour __rcu *next; struct neigh_table *tbl; struct neigh_parms *parms; struct net_device *dev; @@ -111,6 +111,7 @@ struct neighbour { struct sk_buff_head arp_queue; struct timer_list timer; const struct neigh_ops *ops; + struct rcu_head rcu; u8 primary_key[0]; }; @@ -139,7 +140,7 @@ struct pneigh_entry { */ struct neigh_hash_table { - struct neighbour **hash_buckets; + struct neighbour __rcu **hash_buckets; unsigned int hash_mask; __u32 hash_rnd; struct rcu_head rcu; diff --git a/net/core/neighbour.c b/net/core/neighbour.c index dd8920e4f508..3ffafaa0414c 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -139,10 +139,12 @@ static int neigh_forced_gc(struct neigh_table *tbl) nht = rcu_dereference_protected(tbl->nht, lockdep_is_held(&tbl->lock)); for (i = 0; i <= nht->hash_mask; i++) { - struct neighbour *n, **np; + struct neighbour *n; + struct neighbour __rcu **np; np = &nht->hash_buckets[i]; - while ((n = *np) != NULL) { + while ((n = rcu_dereference_protected(*np, + lockdep_is_held(&tbl->lock))) != NULL) { /* Neighbour record may be discarded if: * - nobody refers to it. * - it is not permanent @@ -150,7 +152,9 @@ static int neigh_forced_gc(struct neigh_table *tbl) write_lock(&n->lock); if (atomic_read(&n->refcnt) == 1 && !(n->nud_state & NUD_PERMANENT)) { - *np = n->next; + rcu_assign_pointer(*np, + rcu_dereference_protected(n->next, + lockdep_is_held(&tbl->lock))); n->dead = 1; shrunk = 1; write_unlock(&n->lock); @@ -208,14 +212,18 @@ static void neigh_flush_dev(struct neigh_table *tbl, struct net_device *dev) lockdep_is_held(&tbl->lock)); for (i = 0; i <= nht->hash_mask; i++) { - struct neighbour *n, **np = &nht->hash_buckets[i]; + struct neighbour *n; + struct neighbour __rcu **np = &nht->hash_buckets[i]; - while ((n = *np) != NULL) { + while ((n = rcu_dereference_protected(*np, + lockdep_is_held(&tbl->lock))) != NULL) { if (dev && n->dev != dev) { np = &n->next; continue; } - *np = n->next; + rcu_assign_pointer(*np, + rcu_dereference_protected(n->next, + lockdep_is_held(&tbl->lock))); write_lock(&n->lock); neigh_del_timer(n); n->dead = 1; @@ -323,7 +331,7 @@ static struct neigh_hash_table *neigh_hash_alloc(unsigned int entries) kfree(ret); return NULL; } - ret->hash_buckets = buckets; + rcu_assign_pointer(ret->hash_buckets, buckets); ret->hash_mask = entries - 1; get_random_bytes(&ret->hash_rnd, sizeof(ret->hash_rnd)); return ret; @@ -362,17 +370,22 @@ static struct neigh_hash_table *neigh_hash_grow(struct neigh_table *tbl, for (i = 0; i <= old_nht->hash_mask; i++) { struct neighbour *n, *next; - for (n = old_nht->hash_buckets[i]; + for (n = rcu_dereference_protected(old_nht->hash_buckets[i], + lockdep_is_held(&tbl->lock)); n != NULL; n = next) { hash = tbl->hash(n->primary_key, n->dev, new_nht->hash_rnd); hash &= new_nht->hash_mask; - next = n->next; - - n->next = new_nht->hash_buckets[hash]; - new_nht->hash_buckets[hash] = n; + next = rcu_dereference_protected(n->next, + lockdep_is_held(&tbl->lock)); + + rcu_assign_pointer(n->next, + rcu_dereference_protected( + new_nht->hash_buckets[hash], + lockdep_is_held(&tbl->lock))); + rcu_assign_pointer(new_nht->hash_buckets[hash], n); } } @@ -394,15 +407,18 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey, rcu_read_lock_bh(); nht = rcu_dereference_bh(tbl->nht); hash_val = tbl->hash(pkey, dev, nht->hash_rnd) & nht->hash_mask; - read_lock(&tbl->lock); - for (n = nht->hash_buckets[hash_val]; n; n = n->next) { + + for (n = rcu_dereference_bh(nht->hash_buckets[hash_val]); + n != NULL; + n = rcu_dereference_bh(n->next)) { if (dev == n->dev && !memcmp(n->primary_key, pkey, key_len)) { - neigh_hold(n); + if (!atomic_inc_not_zero(&n->refcnt)) + n = NULL; NEIGH_CACHE_STAT_INC(tbl, hits); break; } } - read_unlock(&tbl->lock); + rcu_read_unlock_bh(); return n; } @@ -421,16 +437,19 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net, rcu_read_lock_bh(); nht = rcu_dereference_bh(tbl->nht); hash_val = tbl->hash(pkey, NULL, nht->hash_rnd) & nht->hash_mask; - read_lock(&tbl->lock); - for (n = nht->hash_buckets[hash_val]; n; n = n->next) { + + for (n = rcu_dereference_bh(nht->hash_buckets[hash_val]); + n != NULL; + n = rcu_dereference_bh(n->next)) { if (!memcmp(n->primary_key, pkey, key_len) && net_eq(dev_net(n->dev), net)) { - neigh_hold(n); + if (!atomic_inc_not_zero(&n->refcnt)) + n = NULL; NEIGH_CACHE_STAT_INC(tbl, hits); break; } } - read_unlock(&tbl->lock); + rcu_read_unlock_bh(); return n; } @@ -483,7 +502,11 @@ struct neighbour *neigh_create(struct neigh_table *tbl, const void *pkey, goto out_tbl_unlock; } - for (n1 = nht->hash_buckets[hash_val]; n1; n1 = n1->next) { + for (n1 = rcu_dereference_protected(nht->hash_buckets[hash_val], + lockdep_is_held(&tbl->lock)); + n1 != NULL; + n1 = rcu_dereference_protected(n1->next, + lockdep_is_held(&tbl->lock))) { if (dev == n1->dev && !memcmp(n1->primary_key, pkey, key_len)) { neigh_hold(n1); rc = n1; @@ -491,10 +514,12 @@ struct neighbour *neigh_create(struct neigh_table *tbl, const void *pkey, } } - n->next = nht->hash_buckets[hash_val]; - nht->hash_buckets[hash_val] = n; n->dead = 0; neigh_hold(n); + rcu_assign_pointer(n->next, + rcu_dereference_protected(nht->hash_buckets[hash_val], + lockdep_is_held(&tbl->lock))); + rcu_assign_pointer(nht->hash_buckets[hash_val], n); write_unlock_bh(&tbl->lock); NEIGH_PRINTK2("neigh %p is created.\n", n); rc = n; @@ -651,6 +676,12 @@ static inline void neigh_parms_put(struct neigh_parms *parms) neigh_parms_destroy(parms); } +static void neigh_destroy_rcu(struct rcu_head *head) +{ + struct neighbour *neigh = container_of(head, struct neighbour, rcu); + + kmem_cache_free(neigh->tbl->kmem_cachep, neigh); +} /* * neighbour must already be out of the table; * @@ -690,7 +721,7 @@ void neigh_destroy(struct neighbour *neigh) NEIGH_PRINTK2("neigh %p is destroyed.\n", neigh); atomic_dec(&neigh->tbl->entries); - kmem_cache_free(neigh->tbl->kmem_cachep, neigh); + call_rcu(&neigh->rcu, neigh_destroy_rcu); } EXPORT_SYMBOL(neigh_destroy); @@ -731,7 +762,8 @@ static void neigh_connect(struct neighbour *neigh) static void neigh_periodic_work(struct work_struct *work) { struct neigh_table *tbl = container_of(work, struct neigh_table, gc_work.work); - struct neighbour *n, **np; + struct neighbour *n; + struct neighbour __rcu **np; unsigned int i; struct neigh_hash_table *nht; @@ -756,7 +788,8 @@ static void neigh_periodic_work(struct work_struct *work) for (i = 0 ; i <= nht->hash_mask; i++) { np = &nht->hash_buckets[i]; - while ((n = *np) != NULL) { + while ((n = rcu_dereference_protected(*np, + lockdep_is_held(&tbl->lock))) != NULL) { unsigned int state; write_lock(&n->lock); @@ -1213,8 +1246,8 @@ static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst, } /* This function can be used in contexts, where only old dev_queue_xmit - worked, f.e. if you want to override normal output path (eql, shaper), - but resolution is not made yet. + * worked, f.e. if you want to override normal output path (eql, shaper), + * but resolution is not made yet. */ int neigh_compat_output(struct sk_buff *skb) @@ -2123,7 +2156,7 @@ static void neigh_update_notify(struct neighbour *neigh) static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, struct netlink_callback *cb) { - struct net * net = sock_net(skb->sk); + struct net *net = sock_net(skb->sk); struct neighbour *n; int rc, h, s_h = cb->args[1]; int idx, s_idx = idx = cb->args[2]; @@ -2132,13 +2165,14 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, rcu_read_lock_bh(); nht = rcu_dereference_bh(tbl->nht); - read_lock(&tbl->lock); for (h = 0; h <= nht->hash_mask; h++) { if (h < s_h) continue; if (h > s_h) s_idx = 0; - for (n = nht->hash_buckets[h], idx = 0; n; n = n->next) { + for (n = rcu_dereference_bh(nht->hash_buckets[h]), idx = 0; + n != NULL; + n = rcu_dereference_bh(n->next)) { if (!net_eq(dev_net(n->dev), net)) continue; if (idx < s_idx) @@ -2150,13 +2184,12 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, rc = -1; goto out; } - next: +next: idx++; } } rc = skb->len; out: - read_unlock(&tbl->lock); rcu_read_unlock_bh(); cb->args[1] = h; cb->args[2] = idx; @@ -2195,11 +2228,13 @@ void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void rcu_read_lock_bh(); nht = rcu_dereference_bh(tbl->nht); - read_lock(&tbl->lock); + read_lock(&tbl->lock); /* avoid resizes */ for (chain = 0; chain <= nht->hash_mask; chain++) { struct neighbour *n; - for (n = nht->hash_buckets[chain]; n; n = n->next) + for (n = rcu_dereference_bh(nht->hash_buckets[chain]); + n != NULL; + n = rcu_dereference_bh(n->next)) cb(n, cookie); } read_unlock(&tbl->lock); @@ -2217,16 +2252,20 @@ void __neigh_for_each_release(struct neigh_table *tbl, nht = rcu_dereference_protected(tbl->nht, lockdep_is_held(&tbl->lock)); for (chain = 0; chain <= nht->hash_mask; chain++) { - struct neighbour *n, **np; + struct neighbour *n; + struct neighbour __rcu **np; np = &nht->hash_buckets[chain]; - while ((n = *np) != NULL) { + while ((n = rcu_dereference_protected(*np, + lockdep_is_held(&tbl->lock))) != NULL) { int release; write_lock(&n->lock); release = cb(n); if (release) { - *np = n->next; + rcu_assign_pointer(*np, + rcu_dereference_protected(n->next, + lockdep_is_held(&tbl->lock))); n->dead = 1; } else np = &n->next; @@ -2250,7 +2289,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq) state->flags &= ~NEIGH_SEQ_IS_PNEIGH; for (bucket = 0; bucket <= nht->hash_mask; bucket++) { - n = nht->hash_buckets[bucket]; + n = rcu_dereference_bh(nht->hash_buckets[bucket]); while (n) { if (!net_eq(dev_net(n->dev), net)) @@ -2267,8 +2306,8 @@ static struct neighbour *neigh_get_first(struct seq_file *seq) break; if (n->nud_state & ~NUD_NOARP) break; - next: - n = n->next; +next: + n = rcu_dereference_bh(n->next); } if (n) @@ -2292,7 +2331,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, if (v) return n; } - n = n->next; + n = rcu_dereference_bh(n->next); while (1) { while (n) { @@ -2309,8 +2348,8 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, if (n->nud_state & ~NUD_NOARP) break; - next: - n = n->next; +next: + n = rcu_dereference_bh(n->next); } if (n) @@ -2319,7 +2358,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, if (++state->bucket > nht->hash_mask) break; - n = nht->hash_buckets[state->bucket]; + n = rcu_dereference_bh(nht->hash_buckets[state->bucket]); } if (n && pos) @@ -2417,7 +2456,6 @@ static void *neigh_get_idx_any(struct seq_file *seq, loff_t *pos) } void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl, unsigned int neigh_seq_flags) - __acquires(tbl->lock) __acquires(rcu_bh) { struct neigh_seq_state *state = seq->private; @@ -2428,7 +2466,7 @@ void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl rcu_read_lock_bh(); state->nht = rcu_dereference_bh(tbl->nht); - read_lock(&tbl->lock); + return *pos ? neigh_get_idx_any(seq, pos) : SEQ_START_TOKEN; } EXPORT_SYMBOL(neigh_seq_start); @@ -2461,13 +2499,8 @@ out: EXPORT_SYMBOL(neigh_seq_next); void neigh_seq_stop(struct seq_file *seq, void *v) - __releases(tbl->lock) __releases(rcu_bh) { - struct neigh_seq_state *state = seq->private; - struct neigh_table *tbl = state->tbl; - - read_unlock(&tbl->lock); rcu_read_unlock_bh(); } EXPORT_SYMBOL(neigh_seq_stop); -- cgit v1.2.3 From 3d3211ef5cf7558d289d1a1efbef8c45595639aa Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Wed, 6 Oct 2010 23:35:15 -0700 Subject: net: netif_set_real_num_rx_queues may cap num_rx_queues at init time Do not set num_rx_queues in netif_set_real_num_rx_queues() some drivers will increase the real_num_rx_queues later due to a feature changes or available interrupts increasing. By setting num_rx_queues here this ends up creating a cap on the number of rx queues available. For example the ixgbe driver sets the max number of queues it intends to use ever then sets the current number in use with the netif_set_num_{rx|tx}_queues calls. With the current implementation the number of rx queues gets limited so when a feature such as DCB or FCoE is enabled the queues are no longer available. kobjects will only be allocated for real_num_rx_queues so the waste in memory is minimal. Signed-off-by: John Fastabend Signed-off-by: Jeff Kirsher Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 7d149550e8d6..fd1b75a47e88 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1593,8 +1593,6 @@ int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq) rxq); if (rc) return rc; - } else { - dev->num_rx_queues = rxq; } dev->real_num_rx_queues = rxq; -- cgit v1.2.3 From 4e7f79511e7332ae4056eda9156a0299511ea41e Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Fri, 8 Oct 2010 10:33:39 -0700 Subject: net: Update kernel-doc for netif_set_real_num_rx_queues() Synchronise the comment with the preceding implementation change. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- net/core/dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index fd1b75a47e88..4962c8afd606 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1576,8 +1576,8 @@ EXPORT_SYMBOL(netif_set_real_num_tx_queues); * * This must be called either with the rtnl_lock held or before * registration of the net device. Returns 0 on success, or a - * negative error code. If called before registration, it also - * sets the maximum number of queues, and always succeeds. + * negative error code. If called before registration, it always + * succeeds. */ int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq) { -- cgit v1.2.3 From 4315d834c1496ddca977e9e22002b77c85bfec2c Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Thu, 7 Oct 2010 10:09:10 +0000 Subject: net: Fix rxq ref counting The rx->count reference is used to track reference counts to the number of rx-queue kobjects created for the device. This patch eliminates initialization of the counter in netif_alloc_rx_queues and instead increments the counter each time a kobject is created. This is now symmetric with the decrement that is done when an object is released. Signed-off-by: Tom Herbert Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 1 - net/core/net-sysfs.c | 2 ++ 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 4962c8afd606..193eafaabd88 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5024,7 +5024,6 @@ static int netif_alloc_rx_queues(struct net_device *dev) return -ENOMEM; } dev->_rx = rx; - atomic_set(&rx->count, count); /* * Set a pointer to first element in the array which holds the diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index fa81fd0a488f..b143173e3eb2 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -726,6 +726,7 @@ static struct kobj_type rx_queue_ktype = { static int rx_queue_add_kobject(struct net_device *net, int index) { struct netdev_rx_queue *queue = net->_rx + index; + struct netdev_rx_queue *first = queue->first; struct kobject *kobj = &queue->kobj; int error = 0; @@ -738,6 +739,7 @@ static int rx_queue_add_kobject(struct net_device *net, int index) } kobject_uevent(kobj, KOBJ_ADD); + atomic_inc(&first->count); return error; } -- cgit v1.2.3 From 34d101dd6204bd100fc2e6f7b5f9a10f959ce2c9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 11 Oct 2010 09:16:57 -0700 Subject: neigh: speedup neigh_hh_init() When a new dst is used to send a frame, neigh_resolve_output() tries to associate an struct hh_cache to this dst, calling neigh_hh_init() with the neigh rwlock write locked. Most of the time, hh_cache is already known and linked into neighbour, so we find it and increment its refcount. This patch changes the logic so that we call neigh_hh_init() with neighbour lock read locked only, so that fast path can be run in parallel by concurrent cpus. This brings part of the speedup we got with commit c7d4426a98a5f (introduce DST_NOCACHE flag) for non cached dsts, even for cached ones, removing one of the contention point that routers hit on multiqueue enabled machines. Further improvements would need to use a seqlock instead of an rwlock to protect neigh->ha[], to not dirty neigh too often and remove two atomic ops. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 6 +++ net/core/dst.c | 4 +- net/core/neighbour.c | 99 +++++++++++++++++++++++++++++------------------ 3 files changed, 69 insertions(+), 40 deletions(-) (limited to 'net/core') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 6abcef67b178..4160db3721ba 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -281,6 +281,12 @@ struct hh_cache { unsigned long hh_data[HH_DATA_ALIGN(LL_MAX_HEADER) / sizeof(long)]; }; +static inline void hh_cache_put(struct hh_cache *hh) +{ + if (atomic_dec_and_test(&hh->hh_refcnt)) + kfree(hh); +} + /* Reserve HH_DATA_MOD byte aligned hard_header_len, but at least that much. * Alternative is: * dev->hard_header_len ? (dev->hard_header_len + diff --git a/net/core/dst.c b/net/core/dst.c index 6c41b1fac3db..978a1ee1f7d0 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -228,8 +228,8 @@ again: child = dst->child; dst->hh = NULL; - if (hh && atomic_dec_and_test(&hh->hh_refcnt)) - kfree(hh); + if (hh) + hh_cache_put(hh); if (neigh) { dst->neighbour = NULL; diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 3ffafaa0414c..2044906ecd1a 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -709,8 +709,7 @@ void neigh_destroy(struct neighbour *neigh) write_seqlock_bh(&hh->hh_lock); hh->hh_output = neigh_blackhole; write_sequnlock_bh(&hh->hh_lock); - if (atomic_dec_and_test(&hh->hh_refcnt)) - kfree(hh); + hh_cache_put(hh); } skb_queue_purge(&neigh->arp_queue); @@ -1210,39 +1209,67 @@ struct neighbour *neigh_event_ns(struct neigh_table *tbl, } EXPORT_SYMBOL(neigh_event_ns); +static inline bool neigh_hh_lookup(struct neighbour *n, struct dst_entry *dst, + __be16 protocol) +{ + struct hh_cache *hh; + + for (hh = n->hh; hh; hh = hh->hh_next) { + if (hh->hh_type == protocol) { + atomic_inc(&hh->hh_refcnt); + if (unlikely(cmpxchg(&dst->hh, NULL, hh) != NULL)) + hh_cache_put(hh); + return true; + } + } + return false; +} + +/* called with read_lock_bh(&n->lock); */ static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst, __be16 protocol) { struct hh_cache *hh; struct net_device *dev = dst->dev; - for (hh = n->hh; hh; hh = hh->hh_next) - if (hh->hh_type == protocol) - break; + if (likely(neigh_hh_lookup(n, dst, protocol))) + return; - if (!hh && (hh = kzalloc(sizeof(*hh), GFP_ATOMIC)) != NULL) { - seqlock_init(&hh->hh_lock); - hh->hh_type = protocol; - atomic_set(&hh->hh_refcnt, 0); - hh->hh_next = NULL; + /* slow path */ + hh = kzalloc(sizeof(*hh), GFP_ATOMIC); + if (!hh) + return; - if (dev->header_ops->cache(n, hh)) { - kfree(hh); - hh = NULL; - } else { - atomic_inc(&hh->hh_refcnt); - hh->hh_next = n->hh; - n->hh = hh; - if (n->nud_state & NUD_CONNECTED) - hh->hh_output = n->ops->hh_output; - else - hh->hh_output = n->ops->output; - } + seqlock_init(&hh->hh_lock); + hh->hh_type = protocol; + atomic_set(&hh->hh_refcnt, 2); + + if (dev->header_ops->cache(n, hh)) { + kfree(hh); + return; } - if (hh) { - atomic_inc(&hh->hh_refcnt); - dst->hh = hh; + read_unlock(&n->lock); + write_lock(&n->lock); + + /* must check if another thread already did the insert */ + if (neigh_hh_lookup(n, dst, protocol)) { + kfree(hh); + goto end; } + + if (n->nud_state & NUD_CONNECTED) + hh->hh_output = n->ops->hh_output; + else + hh->hh_output = n->ops->output; + + hh->hh_next = n->hh; + n->hh = hh; + + if (unlikely(cmpxchg(&dst->hh, NULL, hh) != NULL)) + hh_cache_put(hh); +end: + write_unlock(&n->lock); + read_lock(&n->lock); } /* This function can be used in contexts, where only old dev_queue_xmit @@ -1281,21 +1308,17 @@ int neigh_resolve_output(struct sk_buff *skb) if (!neigh_event_send(neigh, skb)) { int err; struct net_device *dev = neigh->dev; + + read_lock_bh(&neigh->lock); if (dev->header_ops->cache && !dst->hh && - !(dst->flags & DST_NOCACHE)) { - write_lock_bh(&neigh->lock); - if (!dst->hh) - neigh_hh_init(neigh, dst, dst->ops->protocol); - err = dev_hard_header(skb, dev, ntohs(skb->protocol), - neigh->ha, NULL, skb->len); - write_unlock_bh(&neigh->lock); - } else { - read_lock_bh(&neigh->lock); - err = dev_hard_header(skb, dev, ntohs(skb->protocol), - neigh->ha, NULL, skb->len); - read_unlock_bh(&neigh->lock); - } + !(dst->flags & DST_NOCACHE)) + neigh_hh_init(neigh, dst, dst->ops->protocol); + + err = dev_hard_header(skb, dev, ntohs(skb->protocol), + neigh->ha, NULL, skb->len); + read_unlock_bh(&neigh->lock); + if (err >= 0) rc = neigh->ops->queue_xmit(skb); else -- cgit v1.2.3 From 0ed8ddf4045fcfcac36bad753dc4046118c603ec Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 7 Oct 2010 10:44:07 +0000 Subject: neigh: Protect neigh->ha[] with a seqlock Add a seqlock in struct neighbour to protect neigh->ha[], and avoid dirtying neighbour in stress situation (many different flows / dsts) Dirtying takes place because of read_lock(&n->lock) and n->used writes. Switching to a seqlock, and writing n->used only on jiffies changes permits less dirtying. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/neighbour.h | 16 +++++++++++++++- net/core/neighbour.c | 47 ++++++++++++++++++++++++++++++----------------- net/ipv4/arp.c | 6 ++---- net/sched/sch_teql.c | 8 ++++---- 4 files changed, 51 insertions(+), 26 deletions(-) (limited to 'net/core') diff --git a/include/net/neighbour.h b/include/net/neighbour.h index a4538d553704..f04e7a2522c5 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -105,6 +105,7 @@ struct neighbour { atomic_t refcnt; atomic_t probes; rwlock_t lock; + seqlock_t ha_lock; unsigned char ha[ALIGN(MAX_ADDR_LEN, sizeof(unsigned long))]; struct hh_cache *hh; int (*output)(struct sk_buff *skb); @@ -302,7 +303,10 @@ static inline void neigh_confirm(struct neighbour *neigh) static inline int neigh_event_send(struct neighbour *neigh, struct sk_buff *skb) { - neigh->used = jiffies; + unsigned long now = ACCESS_ONCE(jiffies); + + if (neigh->used != now) + neigh->used = now; if (!(neigh->nud_state&(NUD_CONNECTED|NUD_DELAY|NUD_PROBE))) return __neigh_event_send(neigh, skb); return 0; @@ -373,4 +377,14 @@ struct neighbour_cb { #define NEIGH_CB(skb) ((struct neighbour_cb *)(skb)->cb) +static inline void neigh_ha_snapshot(char *dst, const struct neighbour *n, + const struct net_device *dev) +{ + unsigned int seq; + + do { + seq = read_seqbegin(&n->ha_lock); + memcpy(dst, n->ha, dev->addr_len); + } while (read_seqretry(&n->ha_lock, seq)); +} #endif diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 2044906ecd1a..b165b96355bf 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -294,6 +294,7 @@ static struct neighbour *neigh_alloc(struct neigh_table *tbl) skb_queue_head_init(&n->arp_queue); rwlock_init(&n->lock); + seqlock_init(&n->ha_lock); n->updated = n->used = now; n->nud_state = NUD_NONE; n->output = neigh_blackhole; @@ -1015,7 +1016,7 @@ out_unlock_bh: } EXPORT_SYMBOL(__neigh_event_send); -static void neigh_update_hhs(struct neighbour *neigh) +static void neigh_update_hhs(const struct neighbour *neigh) { struct hh_cache *hh; void (*update)(struct hh_cache*, const struct net_device*, const unsigned char *) @@ -1151,7 +1152,9 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new, } if (lladdr != neigh->ha) { + write_seqlock(&neigh->ha_lock); memcpy(&neigh->ha, lladdr, dev->addr_len); + write_sequnlock(&neigh->ha_lock); neigh_update_hhs(neigh); if (!(new & NUD_CONNECTED)) neigh->confirmed = jiffies - @@ -1214,6 +1217,7 @@ static inline bool neigh_hh_lookup(struct neighbour *n, struct dst_entry *dst, { struct hh_cache *hh; + smp_rmb(); /* paired with smp_wmb() in neigh_hh_init() */ for (hh = n->hh; hh; hh = hh->hh_next) { if (hh->hh_type == protocol) { atomic_inc(&hh->hh_refcnt); @@ -1248,8 +1252,8 @@ static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst, kfree(hh); return; } - read_unlock(&n->lock); - write_lock(&n->lock); + + write_lock_bh(&n->lock); /* must check if another thread already did the insert */ if (neigh_hh_lookup(n, dst, protocol)) { @@ -1263,13 +1267,13 @@ static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst, hh->hh_output = n->ops->output; hh->hh_next = n->hh; + smp_wmb(); /* paired with smp_rmb() in neigh_hh_lookup() */ n->hh = hh; if (unlikely(cmpxchg(&dst->hh, NULL, hh) != NULL)) hh_cache_put(hh); end: - write_unlock(&n->lock); - read_lock(&n->lock); + write_unlock_bh(&n->lock); } /* This function can be used in contexts, where only old dev_queue_xmit @@ -1308,16 +1312,18 @@ int neigh_resolve_output(struct sk_buff *skb) if (!neigh_event_send(neigh, skb)) { int err; struct net_device *dev = neigh->dev; + unsigned int seq; - read_lock_bh(&neigh->lock); if (dev->header_ops->cache && !dst->hh && !(dst->flags & DST_NOCACHE)) neigh_hh_init(neigh, dst, dst->ops->protocol); - err = dev_hard_header(skb, dev, ntohs(skb->protocol), - neigh->ha, NULL, skb->len); - read_unlock_bh(&neigh->lock); + do { + seq = read_seqbegin(&neigh->ha_lock); + err = dev_hard_header(skb, dev, ntohs(skb->protocol), + neigh->ha, NULL, skb->len); + } while (read_seqretry(&neigh->ha_lock, seq)); if (err >= 0) rc = neigh->ops->queue_xmit(skb); @@ -1344,13 +1350,16 @@ int neigh_connected_output(struct sk_buff *skb) struct dst_entry *dst = skb_dst(skb); struct neighbour *neigh = dst->neighbour; struct net_device *dev = neigh->dev; + unsigned int seq; __skb_pull(skb, skb_network_offset(skb)); - read_lock_bh(&neigh->lock); - err = dev_hard_header(skb, dev, ntohs(skb->protocol), - neigh->ha, NULL, skb->len); - read_unlock_bh(&neigh->lock); + do { + seq = read_seqbegin(&neigh->ha_lock); + err = dev_hard_header(skb, dev, ntohs(skb->protocol), + neigh->ha, NULL, skb->len); + } while (read_seqretry(&neigh->ha_lock, seq)); + if (err >= 0) err = neigh->ops->queue_xmit(skb); else { @@ -2148,10 +2157,14 @@ static int neigh_fill_info(struct sk_buff *skb, struct neighbour *neigh, read_lock_bh(&neigh->lock); ndm->ndm_state = neigh->nud_state; - if ((neigh->nud_state & NUD_VALID) && - nla_put(skb, NDA_LLADDR, neigh->dev->addr_len, neigh->ha) < 0) { - read_unlock_bh(&neigh->lock); - goto nla_put_failure; + if (neigh->nud_state & NUD_VALID) { + char haddr[MAX_ADDR_LEN]; + + neigh_ha_snapshot(haddr, neigh, neigh->dev); + if (nla_put(skb, NDA_LLADDR, neigh->dev->addr_len, haddr) < 0) { + read_unlock_bh(&neigh->lock); + goto nla_put_failure; + } } ci.ndm_used = jiffies_to_clock_t(now - neigh->used); diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index f35309578170..d8e540c5b071 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -502,10 +502,8 @@ int arp_find(unsigned char *haddr, struct sk_buff *skb) if (n) { n->used = jiffies; - if (n->nud_state&NUD_VALID || neigh_event_send(n, skb) == 0) { - read_lock_bh(&n->lock); - memcpy(haddr, n->ha, dev->addr_len); - read_unlock_bh(&n->lock); + if (n->nud_state & NUD_VALID || neigh_event_send(n, skb) == 0) { + neigh_ha_snapshot(haddr, n, dev); neigh_release(n); return 0; } diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index feaabc103ce6..401af9596709 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -241,11 +241,11 @@ __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device * } if (neigh_event_send(n, skb_res) == 0) { int err; + char haddr[MAX_ADDR_LEN]; - read_lock(&n->lock); - err = dev_hard_header(skb, dev, ntohs(skb->protocol), - n->ha, NULL, skb->len); - read_unlock(&n->lock); + neigh_ha_snapshot(haddr, n, dev); + err = dev_hard_header(skb, dev, ntohs(skb->protocol), haddr, + NULL, skb->len); if (err < 0) { neigh_release(n); -- cgit v1.2.3 From fc66f95c68b6d4535a0ea2ea15d5cf626e310956 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 8 Oct 2010 06:37:34 +0000 Subject: net dst: use a percpu_counter to track entries struct dst_ops tracks number of allocated dst in an atomic_t field, subject to high cache line contention in stress workload. Switch to a percpu_counter, to reduce number of time we need to dirty a central location. Place it on a separate cache line to avoid dirtying read only fields. Stress test : (Sending 160.000.000 UDP frames, IP route cache disabled, dual E5540 @2.53GHz, 32bit kernel, FIB_TRIE, SLUB/NUMA) Before: real 0m51.179s user 0m15.329s sys 10m15.942s After: real 0m45.570s user 0m15.525s sys 9m56.669s With a small reordering of struct neighbour fields, subject of a following patch, (to separate refcnt from other read mostly fields) real 0m41.841s user 0m15.261s sys 8m45.949s Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/dst_ops.h | 37 ++++++++++++++++++++++++++++++++++++- net/bridge/br_netfilter.c | 11 +++++++++-- net/core/dst.c | 6 +++--- net/decnet/dn_route.c | 3 ++- net/ipv4/route.c | 36 ++++++++++++++++++++++-------------- net/ipv4/xfrm4_policy.c | 4 ++-- net/ipv6/route.c | 28 ++++++++++++++++++++-------- net/ipv6/xfrm6_policy.c | 10 ++++++---- 8 files changed, 100 insertions(+), 35 deletions(-) (limited to 'net/core') diff --git a/include/net/dst_ops.h b/include/net/dst_ops.h index d1ff9b7e99b8..1fa5306e3e23 100644 --- a/include/net/dst_ops.h +++ b/include/net/dst_ops.h @@ -1,6 +1,7 @@ #ifndef _NET_DST_OPS_H #define _NET_DST_OPS_H #include +#include struct dst_entry; struct kmem_cachep; @@ -22,7 +23,41 @@ struct dst_ops { void (*update_pmtu)(struct dst_entry *dst, u32 mtu); int (*local_out)(struct sk_buff *skb); - atomic_t entries; struct kmem_cache *kmem_cachep; + + struct percpu_counter pcpuc_entries ____cacheline_aligned_in_smp; }; + +static inline int dst_entries_get_fast(struct dst_ops *dst) +{ + return percpu_counter_read_positive(&dst->pcpuc_entries); +} + +static inline int dst_entries_get_slow(struct dst_ops *dst) +{ + int res; + + local_bh_disable(); + res = percpu_counter_sum_positive(&dst->pcpuc_entries); + local_bh_enable(); + return res; +} + +static inline void dst_entries_add(struct dst_ops *dst, int val) +{ + local_bh_disable(); + percpu_counter_add(&dst->pcpuc_entries, val); + local_bh_enable(); +} + +static inline int dst_entries_init(struct dst_ops *dst) +{ + return percpu_counter_init(&dst->pcpuc_entries, 0); +} + +static inline void dst_entries_destroy(struct dst_ops *dst) +{ + percpu_counter_destroy(&dst->pcpuc_entries); +} + #endif diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index 77f7b5fda45a..7f9ce9600ef3 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -106,7 +106,6 @@ static struct dst_ops fake_dst_ops = { .family = AF_INET, .protocol = cpu_to_be16(ETH_P_IP), .update_pmtu = fake_update_pmtu, - .entries = ATOMIC_INIT(0), }; /* @@ -1003,15 +1002,22 @@ int __init br_netfilter_init(void) { int ret; - ret = nf_register_hooks(br_nf_ops, ARRAY_SIZE(br_nf_ops)); + ret = dst_entries_init(&fake_dst_ops); if (ret < 0) return ret; + + ret = nf_register_hooks(br_nf_ops, ARRAY_SIZE(br_nf_ops)); + if (ret < 0) { + dst_entries_destroy(&fake_dst_ops); + return ret; + } #ifdef CONFIG_SYSCTL brnf_sysctl_header = register_sysctl_paths(brnf_path, brnf_table); if (brnf_sysctl_header == NULL) { printk(KERN_WARNING "br_netfilter: can't register to sysctl.\n"); nf_unregister_hooks(br_nf_ops, ARRAY_SIZE(br_nf_ops)); + dst_entries_destroy(&fake_dst_ops); return -ENOMEM; } #endif @@ -1025,4 +1031,5 @@ void br_netfilter_fini(void) #ifdef CONFIG_SYSCTL unregister_sysctl_table(brnf_sysctl_header); #endif + dst_entries_destroy(&fake_dst_ops); } diff --git a/net/core/dst.c b/net/core/dst.c index 978a1ee1f7d0..32e542d7f472 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -168,7 +168,7 @@ void *dst_alloc(struct dst_ops *ops) { struct dst_entry *dst; - if (ops->gc && atomic_read(&ops->entries) > ops->gc_thresh) { + if (ops->gc && dst_entries_get_fast(ops) > ops->gc_thresh) { if (ops->gc(ops)) return NULL; } @@ -183,7 +183,7 @@ void *dst_alloc(struct dst_ops *ops) #if RT_CACHE_DEBUG >= 2 atomic_inc(&dst_total); #endif - atomic_inc(&ops->entries); + dst_entries_add(ops, 1); return dst; } EXPORT_SYMBOL(dst_alloc); @@ -236,7 +236,7 @@ again: neigh_release(neigh); } - atomic_dec(&dst->ops->entries); + dst_entries_add(dst->ops, -1); if (dst->ops->destroy) dst->ops->destroy(dst); diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 6585ea6d1182..df0f3e54ff8a 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -132,7 +132,6 @@ static struct dst_ops dn_dst_ops = { .negative_advice = dn_dst_negative_advice, .link_failure = dn_dst_link_failure, .update_pmtu = dn_dst_update_pmtu, - .entries = ATOMIC_INIT(0), }; static __inline__ unsigned dn_hash(__le16 src, __le16 dst) @@ -1758,6 +1757,7 @@ void __init dn_route_init(void) dn_dst_ops.kmem_cachep = kmem_cache_create("dn_dst_cache", sizeof(struct dn_route), 0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); + dst_entries_init(&dn_dst_ops); setup_timer(&dn_route_timer, dn_dst_check_expire, 0); dn_route_timer.expires = jiffies + decnet_dst_gc_interval * HZ; add_timer(&dn_route_timer); @@ -1816,5 +1816,6 @@ void __exit dn_route_cleanup(void) dn_run_flush(0); proc_net_remove(&init_net, "decnet_cache"); + dst_entries_destroy(&dn_dst_ops); } diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 3888f6ba0a5c..0755aa4af86c 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -159,7 +159,6 @@ static struct dst_ops ipv4_dst_ops = { .link_failure = ipv4_link_failure, .update_pmtu = ip_rt_update_pmtu, .local_out = __ip_local_out, - .entries = ATOMIC_INIT(0), }; #define ECN_OR_COST(class) TC_PRIO_##class @@ -466,7 +465,7 @@ static int rt_cpu_seq_show(struct seq_file *seq, void *v) seq_printf(seq,"%08x %08x %08x %08x %08x %08x %08x %08x " " %08x %08x %08x %08x %08x %08x %08x %08x %08x \n", - atomic_read(&ipv4_dst_ops.entries), + dst_entries_get_slow(&ipv4_dst_ops), st->in_hit, st->in_slow_tot, st->in_slow_mc, @@ -945,6 +944,7 @@ static int rt_garbage_collect(struct dst_ops *ops) struct rtable *rth, **rthp; unsigned long now = jiffies; int goal; + int entries = dst_entries_get_fast(&ipv4_dst_ops); /* * Garbage collection is pretty expensive, @@ -954,28 +954,28 @@ static int rt_garbage_collect(struct dst_ops *ops) RT_CACHE_STAT_INC(gc_total); if (now - last_gc < ip_rt_gc_min_interval && - atomic_read(&ipv4_dst_ops.entries) < ip_rt_max_size) { + entries < ip_rt_max_size) { RT_CACHE_STAT_INC(gc_ignored); goto out; } + entries = dst_entries_get_slow(&ipv4_dst_ops); /* Calculate number of entries, which we want to expire now. */ - goal = atomic_read(&ipv4_dst_ops.entries) - - (ip_rt_gc_elasticity << rt_hash_log); + goal = entries - (ip_rt_gc_elasticity << rt_hash_log); if (goal <= 0) { if (equilibrium < ipv4_dst_ops.gc_thresh) equilibrium = ipv4_dst_ops.gc_thresh; - goal = atomic_read(&ipv4_dst_ops.entries) - equilibrium; + goal = entries - equilibrium; if (goal > 0) { equilibrium += min_t(unsigned int, goal >> 1, rt_hash_mask + 1); - goal = atomic_read(&ipv4_dst_ops.entries) - equilibrium; + goal = entries - equilibrium; } } else { /* We are in dangerous area. Try to reduce cache really * aggressively. */ goal = max_t(unsigned int, goal >> 1, rt_hash_mask + 1); - equilibrium = atomic_read(&ipv4_dst_ops.entries) - goal; + equilibrium = entries - goal; } if (now - last_gc >= ip_rt_gc_min_interval) @@ -1032,14 +1032,16 @@ static int rt_garbage_collect(struct dst_ops *ops) expire >>= 1; #if RT_CACHE_DEBUG >= 2 printk(KERN_DEBUG "expire>> %u %d %d %d\n", expire, - atomic_read(&ipv4_dst_ops.entries), goal, i); + dst_entries_get_fast(&ipv4_dst_ops), goal, i); #endif - if (atomic_read(&ipv4_dst_ops.entries) < ip_rt_max_size) + if (dst_entries_get_fast(&ipv4_dst_ops) < ip_rt_max_size) goto out; } while (!in_softirq() && time_before_eq(jiffies, now)); - if (atomic_read(&ipv4_dst_ops.entries) < ip_rt_max_size) + if (dst_entries_get_fast(&ipv4_dst_ops) < ip_rt_max_size) + goto out; + if (dst_entries_get_slow(&ipv4_dst_ops) < ip_rt_max_size) goto out; if (net_ratelimit()) printk(KERN_WARNING "dst cache overflow\n"); @@ -1049,11 +1051,12 @@ static int rt_garbage_collect(struct dst_ops *ops) work_done: expire += ip_rt_gc_min_interval; if (expire > ip_rt_gc_timeout || - atomic_read(&ipv4_dst_ops.entries) < ipv4_dst_ops.gc_thresh) + dst_entries_get_fast(&ipv4_dst_ops) < ipv4_dst_ops.gc_thresh || + dst_entries_get_slow(&ipv4_dst_ops) < ipv4_dst_ops.gc_thresh) expire = ip_rt_gc_timeout; #if RT_CACHE_DEBUG >= 2 printk(KERN_DEBUG "expire++ %u %d %d %d\n", expire, - atomic_read(&ipv4_dst_ops.entries), goal, rover); + dst_entries_get_fast(&ipv4_dst_ops), goal, rover); #endif out: return 0; } @@ -2717,7 +2720,6 @@ static struct dst_ops ipv4_dst_blackhole_ops = { .destroy = ipv4_dst_destroy, .check = ipv4_blackhole_dst_check, .update_pmtu = ipv4_rt_blackhole_update_pmtu, - .entries = ATOMIC_INIT(0), }; @@ -3287,6 +3289,12 @@ int __init ip_rt_init(void) ipv4_dst_blackhole_ops.kmem_cachep = ipv4_dst_ops.kmem_cachep; + if (dst_entries_init(&ipv4_dst_ops) < 0) + panic("IP: failed to allocate ipv4_dst_ops counter\n"); + + if (dst_entries_init(&ipv4_dst_blackhole_ops) < 0) + panic("IP: failed to allocate ipv4_dst_blackhole_ops counter\n"); + rt_hash_table = (struct rt_hash_bucket *) alloc_large_system_hash("IP route cache", sizeof(struct rt_hash_bucket), diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index a580349f0b8a..4464f3bff6a7 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -174,7 +174,7 @@ static inline int xfrm4_garbage_collect(struct dst_ops *ops) struct net *net = container_of(ops, struct net, xfrm.xfrm4_dst_ops); xfrm4_policy_afinfo.garbage_collect(net); - return (atomic_read(&ops->entries) > ops->gc_thresh * 2); + return (dst_entries_get_slow(ops) > ops->gc_thresh * 2); } static void xfrm4_update_pmtu(struct dst_entry *dst, u32 mtu) @@ -232,7 +232,6 @@ static struct dst_ops xfrm4_dst_ops = { .ifdown = xfrm4_dst_ifdown, .local_out = __ip_local_out, .gc_thresh = 1024, - .entries = ATOMIC_INIT(0), }; static struct xfrm_policy_afinfo xfrm4_policy_afinfo = { @@ -288,6 +287,7 @@ void __init xfrm4_init(int rt_max_size) * and start cleaning when were 1/2 full */ xfrm4_dst_ops.gc_thresh = rt_max_size/2; + dst_entries_init(&xfrm4_dst_ops); xfrm4_state_init(); xfrm4_policy_init(); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 17e217933885..25661f968f3f 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -109,7 +109,6 @@ static struct dst_ops ip6_dst_ops_template = { .link_failure = ip6_link_failure, .update_pmtu = ip6_rt_update_pmtu, .local_out = __ip6_local_out, - .entries = ATOMIC_INIT(0), }; static void ip6_rt_blackhole_update_pmtu(struct dst_entry *dst, u32 mtu) @@ -122,7 +121,6 @@ static struct dst_ops ip6_dst_blackhole_ops = { .destroy = ip6_dst_destroy, .check = ip6_dst_check, .update_pmtu = ip6_rt_blackhole_update_pmtu, - .entries = ATOMIC_INIT(0), }; static struct rt6_info ip6_null_entry_template = { @@ -1058,19 +1056,22 @@ static int ip6_dst_gc(struct dst_ops *ops) int rt_elasticity = net->ipv6.sysctl.ip6_rt_gc_elasticity; int rt_gc_timeout = net->ipv6.sysctl.ip6_rt_gc_timeout; unsigned long rt_last_gc = net->ipv6.ip6_rt_last_gc; + int entries; + entries = dst_entries_get_fast(ops); if (time_after(rt_last_gc + rt_min_interval, now) && - atomic_read(&ops->entries) <= rt_max_size) + entries <= rt_max_size) goto out; net->ipv6.ip6_rt_gc_expire++; fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net); net->ipv6.ip6_rt_last_gc = now; - if (atomic_read(&ops->entries) < ops->gc_thresh) + entries = dst_entries_get_slow(ops); + if (entries < ops->gc_thresh) net->ipv6.ip6_rt_gc_expire = rt_gc_timeout>>1; out: net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>rt_elasticity; - return atomic_read(&ops->entries) > rt_max_size; + return entries > rt_max_size; } /* Clean host part of a prefix. Not necessary in radix tree, @@ -2524,7 +2525,7 @@ static int rt6_stats_seq_show(struct seq_file *seq, void *v) net->ipv6.rt6_stats->fib_rt_alloc, net->ipv6.rt6_stats->fib_rt_entries, net->ipv6.rt6_stats->fib_rt_cache, - atomic_read(&net->ipv6.ip6_dst_ops.entries), + dst_entries_get_slow(&net->ipv6.ip6_dst_ops), net->ipv6.rt6_stats->fib_discarded_routes); return 0; @@ -2666,11 +2667,14 @@ static int __net_init ip6_route_net_init(struct net *net) memcpy(&net->ipv6.ip6_dst_ops, &ip6_dst_ops_template, sizeof(net->ipv6.ip6_dst_ops)); + if (dst_entries_init(&net->ipv6.ip6_dst_ops) < 0) + goto out_ip6_dst_ops; + net->ipv6.ip6_null_entry = kmemdup(&ip6_null_entry_template, sizeof(*net->ipv6.ip6_null_entry), GFP_KERNEL); if (!net->ipv6.ip6_null_entry) - goto out_ip6_dst_ops; + goto out_ip6_dst_entries; net->ipv6.ip6_null_entry->dst.path = (struct dst_entry *)net->ipv6.ip6_null_entry; net->ipv6.ip6_null_entry->dst.ops = &net->ipv6.ip6_dst_ops; @@ -2720,6 +2724,8 @@ out_ip6_prohibit_entry: out_ip6_null_entry: kfree(net->ipv6.ip6_null_entry); #endif +out_ip6_dst_entries: + dst_entries_destroy(&net->ipv6.ip6_dst_ops); out_ip6_dst_ops: goto out; } @@ -2758,10 +2764,14 @@ int __init ip6_route_init(void) if (!ip6_dst_ops_template.kmem_cachep) goto out; - ret = register_pernet_subsys(&ip6_route_net_ops); + ret = dst_entries_init(&ip6_dst_blackhole_ops); if (ret) goto out_kmem_cache; + ret = register_pernet_subsys(&ip6_route_net_ops); + if (ret) + goto out_dst_entries; + ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops_template.kmem_cachep; /* Registering of the loopback is done before this portion of code, @@ -2808,6 +2818,8 @@ out_fib6_init: fib6_gc_cleanup(); out_register_subsys: unregister_pernet_subsys(&ip6_route_net_ops); +out_dst_entries: + dst_entries_destroy(&ip6_dst_blackhole_ops); out_kmem_cache: kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep); goto out; diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 39676eac3a37..7e74023ea6e4 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -199,7 +199,7 @@ static inline int xfrm6_garbage_collect(struct dst_ops *ops) struct net *net = container_of(ops, struct net, xfrm.xfrm6_dst_ops); xfrm6_policy_afinfo.garbage_collect(net); - return atomic_read(&ops->entries) > ops->gc_thresh * 2; + return dst_entries_get_fast(ops) > ops->gc_thresh * 2; } static void xfrm6_update_pmtu(struct dst_entry *dst, u32 mtu) @@ -255,7 +255,6 @@ static struct dst_ops xfrm6_dst_ops = { .ifdown = xfrm6_dst_ifdown, .local_out = __ip6_local_out, .gc_thresh = 1024, - .entries = ATOMIC_INIT(0), }; static struct xfrm_policy_afinfo xfrm6_policy_afinfo = { @@ -312,11 +311,13 @@ int __init xfrm6_init(void) */ gc_thresh = FIB6_TABLE_HASHSZ * 8; xfrm6_dst_ops.gc_thresh = (gc_thresh < 1024) ? 1024 : gc_thresh; + dst_entries_init(&xfrm6_dst_ops); ret = xfrm6_policy_init(); - if (ret) + if (ret) { + dst_entries_destroy(&xfrm6_dst_ops); goto out; - + } ret = xfrm6_state_init(); if (ret) goto out_policy; @@ -341,4 +342,5 @@ void xfrm6_fini(void) //xfrm6_input_fini(); xfrm6_policy_fini(); xfrm6_state_fini(); + dst_entries_destroy(&xfrm6_dst_ops); } -- cgit v1.2.3 From 29b4433d991c88d86ca48a4c1cc33c671475be4b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 11 Oct 2010 10:22:12 +0000 Subject: net: percpu net_device refcount We tried very hard to remove all possible dev_hold()/dev_put() pairs in network stack, using RCU conversions. There is still an unavoidable device refcount change for every dst we create/destroy, and this can slow down some workloads (routers or some app servers, mmap af_packet) We can switch to a percpu refcount implementation, now dynamic per_cpu infrastructure is mature. On a 64 cpus machine, this consumes 256 bytes per device. On x86, dev_hold(dev) code : before lock incl 0x280(%ebx) after: movl 0x260(%ebx),%eax incl fs:(%eax) Stress bench : (Sending 160.000.000 UDP frames, IP route cache disabled, dual E5540 @2.53GHz, 32bit kernel, FIB_TRIE) Before: real 1m1.662s user 0m14.373s sys 12m55.960s After: real 0m51.179s user 0m15.329s sys 10m15.942s Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/infiniband/hw/nes/nes_cm.c | 4 ++-- drivers/infiniband/hw/nes/nes_verbs.c | 4 ++-- include/linux/netdevice.h | 7 +++--- net/core/dev.c | 40 +++++++++++++++++++++++++++++------ 4 files changed, 41 insertions(+), 14 deletions(-) (limited to 'net/core') diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c index 61e0efd4ccfb..6220d9d75b58 100644 --- a/drivers/infiniband/hw/nes/nes_cm.c +++ b/drivers/infiniband/hw/nes/nes_cm.c @@ -2701,7 +2701,7 @@ static int nes_disconnect(struct nes_qp *nesqp, int abrupt) nesibdev = nesvnic->nesibdev; nes_debug(NES_DBG_CM, "netdev refcnt = %u.\n", - atomic_read(&nesvnic->netdev->refcnt)); + netdev_refcnt_read(nesvnic->netdev)); if (nesqp->active_conn) { @@ -2791,7 +2791,7 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) atomic_inc(&cm_accepts); nes_debug(NES_DBG_CM, "netdev refcnt = %u.\n", - atomic_read(&nesvnic->netdev->refcnt)); + netdev_refcnt_read(nesvnic->netdev)); /* allocate the ietf frame and space for private data */ nesqp->ietf_frame = pci_alloc_consistent(nesdev->pcidev, diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c index 9046e6675686..546fc22405fe 100644 --- a/drivers/infiniband/hw/nes/nes_verbs.c +++ b/drivers/infiniband/hw/nes/nes_verbs.c @@ -785,7 +785,7 @@ static struct ib_pd *nes_alloc_pd(struct ib_device *ibdev, nes_debug(NES_DBG_PD, "nesvnic=%p, netdev=%p %s, ibdev=%p, context=%p, netdev refcnt=%u\n", nesvnic, nesdev->netdev[0], nesdev->netdev[0]->name, ibdev, context, - atomic_read(&nesvnic->netdev->refcnt)); + netdev_refcnt_read(nesvnic->netdev)); err = nes_alloc_resource(nesadapter, nesadapter->allocated_pds, nesadapter->max_pd, &pd_num, &nesadapter->next_pd); @@ -1416,7 +1416,7 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd, /* update the QP table */ nesdev->nesadapter->qp_table[nesqp->hwqp.qp_id-NES_FIRST_QPN] = nesqp; nes_debug(NES_DBG_QP, "netdev refcnt=%u\n", - atomic_read(&nesvnic->netdev->refcnt)); + netdev_refcnt_read(nesvnic->netdev)); return &nesqp->ibqp; } diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 4160db3721ba..14fbb04c459d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1026,7 +1026,7 @@ struct net_device { struct timer_list watchdog_timer; /* Number of references to this device */ - atomic_t refcnt ____cacheline_aligned_in_smp; + int __percpu *pcpu_refcnt; /* delayed register/unregister */ struct list_head todo_list; @@ -1330,6 +1330,7 @@ static inline void unregister_netdevice(struct net_device *dev) unregister_netdevice_queue(dev, NULL); } +extern int netdev_refcnt_read(const struct net_device *dev); extern void free_netdev(struct net_device *dev); extern void synchronize_net(void); extern int register_netdevice_notifier(struct notifier_block *nb); @@ -1798,7 +1799,7 @@ extern void netdev_run_todo(void); */ static inline void dev_put(struct net_device *dev) { - atomic_dec(&dev->refcnt); + irqsafe_cpu_dec(*dev->pcpu_refcnt); } /** @@ -1809,7 +1810,7 @@ static inline void dev_put(struct net_device *dev) */ static inline void dev_hold(struct net_device *dev) { - atomic_inc(&dev->refcnt); + irqsafe_cpu_inc(*dev->pcpu_refcnt); } /* Carrier loss detection, dial on demand. The functions netif_carrier_on diff --git a/net/core/dev.c b/net/core/dev.c index 193eafaabd88..04972a4783e2 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5192,9 +5192,6 @@ int init_dummy_netdev(struct net_device *dev) */ dev->reg_state = NETREG_DUMMY; - /* initialize the ref count */ - atomic_set(&dev->refcnt, 1); - /* NAPI wants this */ INIT_LIST_HEAD(&dev->napi_list); @@ -5202,6 +5199,11 @@ int init_dummy_netdev(struct net_device *dev) set_bit(__LINK_STATE_PRESENT, &dev->state); set_bit(__LINK_STATE_START, &dev->state); + /* Note : We dont allocate pcpu_refcnt for dummy devices, + * because users of this 'device' dont need to change + * its refcount. + */ + return 0; } EXPORT_SYMBOL_GPL(init_dummy_netdev); @@ -5243,6 +5245,16 @@ out: } EXPORT_SYMBOL(register_netdev); +int netdev_refcnt_read(const struct net_device *dev) +{ + int i, refcnt = 0; + + for_each_possible_cpu(i) + refcnt += *per_cpu_ptr(dev->pcpu_refcnt, i); + return refcnt; +} +EXPORT_SYMBOL(netdev_refcnt_read); + /* * netdev_wait_allrefs - wait until all references are gone. * @@ -5257,11 +5269,14 @@ EXPORT_SYMBOL(register_netdev); static void netdev_wait_allrefs(struct net_device *dev) { unsigned long rebroadcast_time, warning_time; + int refcnt; linkwatch_forget_dev(dev); rebroadcast_time = warning_time = jiffies; - while (atomic_read(&dev->refcnt) != 0) { + refcnt = netdev_refcnt_read(dev); + + while (refcnt != 0) { if (time_after(jiffies, rebroadcast_time + 1 * HZ)) { rtnl_lock(); @@ -5288,11 +5303,13 @@ static void netdev_wait_allrefs(struct net_device *dev) msleep(250); + refcnt = netdev_refcnt_read(dev); + if (time_after(jiffies, warning_time + 10 * HZ)) { printk(KERN_EMERG "unregister_netdevice: " "waiting for %s to become free. Usage " "count = %d\n", - dev->name, atomic_read(&dev->refcnt)); + dev->name, refcnt); warning_time = jiffies; } } @@ -5350,7 +5367,7 @@ void netdev_run_todo(void) netdev_wait_allrefs(dev); /* paranoia */ - BUG_ON(atomic_read(&dev->refcnt)); + BUG_ON(netdev_refcnt_read(dev)); WARN_ON(rcu_dereference_raw(dev->ip_ptr)); WARN_ON(dev->ip6_ptr); WARN_ON(dev->dn_ptr); @@ -5520,9 +5537,13 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, dev = PTR_ALIGN(p, NETDEV_ALIGN); dev->padded = (char *)dev - (char *)p; - if (dev_addr_init(dev)) + dev->pcpu_refcnt = alloc_percpu(int); + if (!dev->pcpu_refcnt) goto free_tx; + if (dev_addr_init(dev)) + goto free_pcpu; + dev_mc_init(dev); dev_uc_init(dev); @@ -5553,6 +5574,8 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, free_tx: kfree(tx); +free_pcpu: + free_percpu(dev->pcpu_refcnt); free_p: kfree(p); return NULL; @@ -5586,6 +5609,9 @@ void free_netdev(struct net_device *dev) list_for_each_entry_safe(p, n, &dev->napi_list, dev_list) netif_napi_del(p); + free_percpu(dev->pcpu_refcnt); + dev->pcpu_refcnt = NULL; + /* Compatibility with error handling in drivers */ if (dev->reg_state == NETREG_UNINITIALIZED) { kfree((char *)dev - dev->padded); -- cgit v1.2.3 From 564824b0c52c34692d804bb6ea214451615b0b50 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 11 Oct 2010 19:05:25 +0000 Subject: net: allocate skbs on local node commit b30973f877 (node-aware skb allocation) spread a wrong habit of allocating net drivers skbs on a given memory node : The one closest to the NIC hardware. This is wrong because as soon as we try to scale network stack, we need to use many cpus to handle traffic and hit slub/slab management on cross-node allocations/frees when these cpus have to alloc/free skbs bound to a central node. skb allocated in RX path are ephemeral, they have a very short lifetime : Extra cost to maintain NUMA affinity is too expensive. What appeared as a nice idea four years ago is in fact a bad one. In 2010, NIC hardwares are multiqueue, or we use RPS to spread the load, and two 10Gb NIC might deliver more than 28 million packets per second, needing all the available cpus. Cost of cross-node handling in network and vm stacks outperforms the small benefit hardware had when doing its DMA transfert in its 'local' memory node at RX time. Even trying to differentiate the two allocations done for one skb (the sk_buff on local node, the data part on NIC hardware node) is not enough to bring good performance. Signed-off-by: Eric Dumazet Acked-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/skbuff.h | 20 ++++++++++++++++---- net/core/skbuff.c | 13 +------------ 2 files changed, 17 insertions(+), 16 deletions(-) (limited to 'net/core') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 0b53c43ac92e..05a358f1ba11 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -496,13 +496,13 @@ extern struct sk_buff *__alloc_skb(unsigned int size, static inline struct sk_buff *alloc_skb(unsigned int size, gfp_t priority) { - return __alloc_skb(size, priority, 0, -1); + return __alloc_skb(size, priority, 0, NUMA_NO_NODE); } static inline struct sk_buff *alloc_skb_fclone(unsigned int size, gfp_t priority) { - return __alloc_skb(size, priority, 1, -1); + return __alloc_skb(size, priority, 1, NUMA_NO_NODE); } extern bool skb_recycle_check(struct sk_buff *skb, int skb_size); @@ -1563,13 +1563,25 @@ static inline struct sk_buff *netdev_alloc_skb_ip_align(struct net_device *dev, return skb; } -extern struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask); +/** + * __netdev_alloc_page - allocate a page for ps-rx on a specific device + * @dev: network device to receive on + * @gfp_mask: alloc_pages_node mask + * + * Allocate a new page. dev currently unused. + * + * %NULL is returned if there is no free memory. + */ +static inline struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask) +{ + return alloc_pages_node(NUMA_NO_NODE, gfp_mask, 0); +} /** * netdev_alloc_page - allocate a page for ps-rx on a specific device * @dev: network device to receive on * - * Allocate a new page node local to the specified device. + * Allocate a new page. dev currently unused. * * %NULL is returned if there is no free memory. */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 752c1972b3a7..4e8b82e167d8 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -247,10 +247,9 @@ EXPORT_SYMBOL(__alloc_skb); struct sk_buff *__netdev_alloc_skb(struct net_device *dev, unsigned int length, gfp_t gfp_mask) { - int node = dev->dev.parent ? dev_to_node(dev->dev.parent) : -1; struct sk_buff *skb; - skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask, 0, node); + skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask, 0, NUMA_NO_NODE); if (likely(skb)) { skb_reserve(skb, NET_SKB_PAD); skb->dev = dev; @@ -259,16 +258,6 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev, } EXPORT_SYMBOL(__netdev_alloc_skb); -struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask) -{ - int node = dev->dev.parent ? dev_to_node(dev->dev.parent) : -1; - struct page *page; - - page = alloc_pages_node(node, gfp_mask, 0); - return page; -} -EXPORT_SYMBOL(__netdev_alloc_page); - void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page, int off, int size) { -- cgit v1.2.3 From a0a4a85a15df6335e3d11f83b2ac06ebebea313f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 13 Oct 2010 04:43:04 +0000 Subject: fib: remove a useless synchronize_rcu() call fib_nl_delrule() calls synchronize_rcu() for no apparent reason, while rtnl is held. I suspect it was done to avoid an atomic_inc_not_zero() in fib_rules_lookup(), which commit 7fa7cb7109d07 added anyway. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/fib_rules.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net/core') diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 21698f8c49ee..1bc3f253ba6c 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -494,7 +494,6 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) } } - synchronize_rcu(); notify_rule_change(RTM_DELRULE, rule, ops, nlh, NETLINK_CB(skb).pid); fib_rule_put(rule); -- cgit v1.2.3 From c2355e1ab910278a94d487b78590ee3c8eecd08a Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Wed, 13 Oct 2010 16:01:49 +0000 Subject: bonding: Fix bonding drivers improper modification of netpoll structure The bonding driver currently modifies the netpoll structure in its xmit path while sending frames from netpoll. This is racy, as other cpus can access the netpoll structure in parallel. Since the bonding driver points np->dev to a slave device, other cpus can inadvertently attempt to send data directly to slave devices, leading to improper locking with the bonding master, lost frames, and deadlocks. This patch fixes that up. This patch also removes the real_dev pointer from the netpoll structure as that data is really only used by bonding in the poll_controller, and we can emulate its behavior by check each slave for IS_UP. Signed-off-by: Neil Horman Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 15 +++++++++------ include/linux/netpoll.h | 9 +++++++-- net/core/netpoll.c | 6 +++--- 3 files changed, 19 insertions(+), 11 deletions(-) (limited to 'net/core') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 7703d35de65d..813cc2f8edd6 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -449,11 +449,9 @@ int bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb, if (unlikely(bond->dev->priv_flags & IFF_IN_NETPOLL)) { struct netpoll *np = bond->dev->npinfo->netpoll; slave_dev->npinfo = bond->dev->npinfo; - np->real_dev = np->dev = skb->dev; slave_dev->priv_flags |= IFF_IN_NETPOLL; - netpoll_send_skb(np, skb); + netpoll_send_skb_on_dev(np, skb, slave_dev); slave_dev->priv_flags &= ~IFF_IN_NETPOLL; - np->dev = bond->dev; } else #endif dev_queue_xmit(skb); @@ -1332,9 +1330,14 @@ static bool slaves_support_netpoll(struct net_device *bond_dev) static void bond_poll_controller(struct net_device *bond_dev) { - struct net_device *dev = bond_dev->npinfo->netpoll->real_dev; - if (dev != bond_dev) - netpoll_poll_dev(dev); + struct bonding *bond = netdev_priv(bond_dev); + struct slave *slave; + int i; + + bond_for_each_slave(bond, slave, i) { + if (slave->dev && IS_UP(slave->dev)) + netpoll_poll_dev(slave->dev); + } } static void bond_netpoll_cleanup(struct net_device *bond_dev) diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 50d8009be86c..79358bb712c6 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -14,7 +14,6 @@ struct netpoll { struct net_device *dev; - struct net_device *real_dev; char dev_name[IFNAMSIZ]; const char *name; void (*rx_hook)(struct netpoll *, int, char *, int); @@ -53,7 +52,13 @@ void netpoll_set_trap(int trap); void __netpoll_cleanup(struct netpoll *np); void netpoll_cleanup(struct netpoll *np); int __netpoll_rx(struct sk_buff *skb); -void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb); +void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, + struct net_device *dev); +static inline void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) +{ + netpoll_send_skb_on_dev(np, skb, np->dev); +} + #ifdef CONFIG_NETPOLL diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 537e01afd81b..4e98ffac3af0 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -288,11 +288,11 @@ static int netpoll_owner_active(struct net_device *dev) return 0; } -void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) +void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, + struct net_device *dev) { int status = NETDEV_TX_BUSY; unsigned long tries; - struct net_device *dev = np->dev; const struct net_device_ops *ops = dev->netdev_ops; /* It is up to the caller to keep npinfo alive. */ struct netpoll_info *npinfo = np->dev->npinfo; @@ -346,7 +346,7 @@ void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) schedule_delayed_work(&npinfo->tx_work,0); } } -EXPORT_SYMBOL(netpoll_send_skb); +EXPORT_SYMBOL(netpoll_send_skb_on_dev); void netpoll_send_udp(struct netpoll *np, const char *msg, int len) { -- cgit v1.2.3 From 990c3d6f9c4115347659fc2b163907c8c832ae44 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Wed, 13 Oct 2010 16:01:51 +0000 Subject: bonding: Fix napi poll for bonding driver Usually the netpoll path, when preforming a napi poll can get away with just polling all the napi instances of the configured device. Thats not the case for the bonding driver however, as the napi instances which may wind up getting flagged as needing polling after the poll_controller call don't belong to the bonded device, but rather to the slave devices. Fix this by checking the device in question for the IFF_MASTER flag, if set, we know we need to check the full poll list for this cpu, rather than just the devices napi instance list. Signed-off-by: Neil Horman Signed-off-by: David S. Miller --- net/core/netpoll.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'net/core') diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 4e98ffac3af0..d79d221fd1f4 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -156,8 +156,15 @@ static void poll_napi(struct net_device *dev) { struct napi_struct *napi; int budget = 16; + struct softnet_data *sd = &__get_cpu_var(softnet_data); + struct list_head *nlist; - list_for_each_entry(napi, &dev->napi_list, dev_list) { + if (dev->flags & IFF_MASTER) + nlist = &sd->poll_list; + else + nlist = &dev->napi_list; + + list_for_each_entry(napi, nlist, dev_list) { if (napi->poll_owner != smp_processor_id() && spin_trylock(&napi->poll_lock)) { budget = poll_one_napi(dev->npinfo, napi, budget); -- cgit v1.2.3 From f13d493d9cf772d510d78ae00bb9f4d680b3170b Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Tue, 19 Oct 2010 07:04:26 +0000 Subject: netpoll: Revert napi_poll fix for bonding driver In an erlier patch I modified napi_poll so that devices with IFF_MASTER polled the per_cpu list instead of the device list for napi. I did this because the bonding driver has no napi instances to poll, it instead expects to check the slave devices napi instances, which napi_poll was unaware of. Looking at this more closely however, I now see this isn't strictly needed. As the bond driver poll_controller calls the slaves poll_controller via netpoll_poll_dev, which recursively calls poll_napi on each slave, allowing those napi instances to get serviced. The earlier patch isn't at all harmfull, its just not needed, so lets revert it to make the code cleaner. Sorry for the noise, Signed-off-by: Neil Horman Reviewed-by: WANG Cong Signed-off-by: David S. Miller --- net/core/netpoll.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) (limited to 'net/core') diff --git a/net/core/netpoll.c b/net/core/netpoll.c index d79d221fd1f4..4e98ffac3af0 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -156,15 +156,8 @@ static void poll_napi(struct net_device *dev) { struct napi_struct *napi; int budget = 16; - struct softnet_data *sd = &__get_cpu_var(softnet_data); - struct list_head *nlist; - if (dev->flags & IFF_MASTER) - nlist = &sd->poll_list; - else - nlist = &dev->napi_list; - - list_for_each_entry(napi, nlist, dev_list) { + list_for_each_entry(napi, &dev->napi_list, dev_list) { if (napi->poll_owner != smp_processor_id() && spin_trylock(&napi->poll_lock)) { budget = poll_one_napi(dev->npinfo, napi, budget); -- cgit v1.2.3 From 55513fb4281464e97aa1ff2b9c906ca5aed917c5 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 18 Oct 2010 17:55:58 +0000 Subject: net: fail alloc_netdev_mq if queue count < 1 In alloc_netdev_mq fail if requested queue_count < 1. Signed-off-by: Tom Herbert Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 04972a4783e2..f44d29ae5d67 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5511,6 +5511,12 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, BUG_ON(strlen(name) >= sizeof(dev->name)); + if (queue_count < 1) { + pr_err("alloc_netdev: Unable to allocate device " + "with zero queues.\n"); + return NULL; + } + alloc_size = sizeof(struct net_device); if (sizeof_priv) { /* ensure 32-byte alignment of private area */ -- cgit v1.2.3 From bd25fa7ba59cd26094319dfba0011b48465f7355 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 18 Oct 2010 18:00:16 +0000 Subject: net: cleanups in RX queue allocation Clean up in RX queue allocation. In netif_set_real_num_rx_queues return error on attempt to set zero queues, or requested number is greater than number of allocated queues. In netif_alloc_rx_queues, do BUG_ON if queue_count is zero. Signed-off-by: Tom Herbert Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 36 +++++++++++++++++------------------- 1 file changed, 17 insertions(+), 19 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index f44d29ae5d67..d33adecec44b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1583,12 +1583,12 @@ int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq) { int rc; + if (rxq < 1 || rxq > dev->num_rx_queues) + return -EINVAL; + if (dev->reg_state == NETREG_REGISTERED) { ASSERT_RTNL(); - if (rxq > dev->num_rx_queues) - return -EINVAL; - rc = net_rx_queue_update_kobjects(dev, dev->real_num_rx_queues, rxq); if (rc) @@ -5013,25 +5013,23 @@ static int netif_alloc_rx_queues(struct net_device *dev) { #ifdef CONFIG_RPS unsigned int i, count = dev->num_rx_queues; + struct netdev_rx_queue *rx; - if (count) { - struct netdev_rx_queue *rx; - - rx = kcalloc(count, sizeof(struct netdev_rx_queue), GFP_KERNEL); - if (!rx) { - pr_err("netdev: Unable to allocate %u rx queues.\n", - count); - return -ENOMEM; - } - dev->_rx = rx; + BUG_ON(count < 1); - /* - * Set a pointer to first element in the array which holds the - * reference count. - */ - for (i = 0; i < count; i++) - rx[i].first = rx; + rx = kcalloc(count, sizeof(struct netdev_rx_queue), GFP_KERNEL); + if (!rx) { + pr_err("netdev: Unable to allocate %u rx queues.\n", count); + return -ENOMEM; } + dev->_rx = rx; + + /* + * Set a pointer to first element in the array which holds the + * reference count. + */ + for (i = 0; i < count; i++) + rx[i].first = rx; #endif return 0; } -- cgit v1.2.3 From e6484930d7c73d324bccda7d43d131088da697b9 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 18 Oct 2010 18:04:39 +0000 Subject: net: allocate tx queues in register_netdevice This patch introduces netif_alloc_netdev_queues which is called from register_device instead of alloc_netdev_mq. This makes TX queue allocation symmetric with RX allocation. Also, queue locks allocation is done in netdev_init_one_queue. Change set_real_num_tx_queues to fail if requested number < 1 or greater than number of allocated queues. Signed-off-by: Tom Herbert Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 +- net/core/dev.c | 106 +++++++++++++++++++++++----------------------- 2 files changed, 55 insertions(+), 55 deletions(-) (limited to 'net/core') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 14fbb04c459d..880d56565828 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1696,8 +1696,8 @@ static inline int netif_is_multiqueue(const struct net_device *dev) return dev->num_tx_queues > 1; } -extern void netif_set_real_num_tx_queues(struct net_device *dev, - unsigned int txq); +extern int netif_set_real_num_tx_queues(struct net_device *dev, + unsigned int txq); #ifdef CONFIG_RPS extern int netif_set_real_num_rx_queues(struct net_device *dev, diff --git a/net/core/dev.c b/net/core/dev.c index d33adecec44b..4c3ac53e4b16 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1553,18 +1553,20 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev) * Routine to help set real_num_tx_queues. To avoid skbs mapped to queues * greater then real_num_tx_queues stale skbs on the qdisc must be flushed. */ -void netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq) +int netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq) { - unsigned int real_num = dev->real_num_tx_queues; + if (txq < 1 || txq > dev->num_tx_queues) + return -EINVAL; - if (unlikely(txq > dev->num_tx_queues)) - ; - else if (txq > real_num) - dev->real_num_tx_queues = txq; - else if (txq < real_num) { - dev->real_num_tx_queues = txq; - qdisc_reset_all_tx_gt(dev, txq); + if (dev->reg_state == NETREG_REGISTERED) { + ASSERT_RTNL(); + + if (txq < dev->real_num_tx_queues) + qdisc_reset_all_tx_gt(dev, txq); } + + dev->real_num_tx_queues = txq; + return 0; } EXPORT_SYMBOL(netif_set_real_num_tx_queues); @@ -4928,20 +4930,6 @@ static void rollback_registered(struct net_device *dev) rollback_registered_many(&single); } -static void __netdev_init_queue_locks_one(struct net_device *dev, - struct netdev_queue *dev_queue, - void *_unused) -{ - spin_lock_init(&dev_queue->_xmit_lock); - netdev_set_xmit_lockdep_class(&dev_queue->_xmit_lock, dev->type); - dev_queue->xmit_lock_owner = -1; -} - -static void netdev_init_queue_locks(struct net_device *dev) -{ - netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL); -} - unsigned long netdev_fix_features(unsigned long features, const char *name) { /* Fix illegal SG+CSUM combinations. */ @@ -5034,6 +5022,41 @@ static int netif_alloc_rx_queues(struct net_device *dev) return 0; } +static int netif_alloc_netdev_queues(struct net_device *dev) +{ + unsigned int count = dev->num_tx_queues; + struct netdev_queue *tx; + + BUG_ON(count < 1); + + tx = kcalloc(count, sizeof(struct netdev_queue), GFP_KERNEL); + if (!tx) { + pr_err("netdev: Unable to allocate %u tx queues.\n", + count); + return -ENOMEM; + } + dev->_tx = tx; + return 0; +} + +static void netdev_init_one_queue(struct net_device *dev, + struct netdev_queue *queue, + void *_unused) +{ + queue->dev = dev; + + /* Initialize queue lock */ + spin_lock_init(&queue->_xmit_lock); + netdev_set_xmit_lockdep_class(&queue->_xmit_lock, dev->type); + queue->xmit_lock_owner = -1; +} + +static void netdev_init_queues(struct net_device *dev) +{ + netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL); + spin_lock_init(&dev->tx_global_lock); +} + /** * register_netdevice - register a network device * @dev: device to register @@ -5067,7 +5090,6 @@ int register_netdevice(struct net_device *dev) spin_lock_init(&dev->addr_list_lock); netdev_set_addr_lockdep_class(dev); - netdev_init_queue_locks(dev); dev->iflink = -1; @@ -5075,6 +5097,12 @@ int register_netdevice(struct net_device *dev) if (ret) goto out; + ret = netif_alloc_netdev_queues(dev); + if (ret) + goto out; + + netdev_init_queues(dev); + /* Init, if this function is available */ if (dev->netdev_ops->ndo_init) { ret = dev->netdev_ops->ndo_init(dev); @@ -5456,19 +5484,6 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev, } EXPORT_SYMBOL(dev_get_stats); -static void netdev_init_one_queue(struct net_device *dev, - struct netdev_queue *queue, - void *_unused) -{ - queue->dev = dev; -} - -static void netdev_init_queues(struct net_device *dev) -{ - netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL); - spin_lock_init(&dev->tx_global_lock); -} - struct netdev_queue *dev_ingress_queue_create(struct net_device *dev) { struct netdev_queue *queue = dev_ingress_queue(dev); @@ -5480,7 +5495,6 @@ struct netdev_queue *dev_ingress_queue_create(struct net_device *dev) if (!queue) return NULL; netdev_init_one_queue(dev, queue, NULL); - __netdev_init_queue_locks_one(dev, queue, NULL); queue->qdisc = &noop_qdisc; queue->qdisc_sleeping = &noop_qdisc; rcu_assign_pointer(dev->ingress_queue, queue); @@ -5502,7 +5516,6 @@ struct netdev_queue *dev_ingress_queue_create(struct net_device *dev) struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, void (*setup)(struct net_device *), unsigned int queue_count) { - struct netdev_queue *tx; struct net_device *dev; size_t alloc_size; struct net_device *p; @@ -5530,20 +5543,12 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, return NULL; } - tx = kcalloc(queue_count, sizeof(struct netdev_queue), GFP_KERNEL); - if (!tx) { - printk(KERN_ERR "alloc_netdev: Unable to allocate " - "tx qdiscs.\n"); - goto free_p; - } - - dev = PTR_ALIGN(p, NETDEV_ALIGN); dev->padded = (char *)dev - (char *)p; dev->pcpu_refcnt = alloc_percpu(int); if (!dev->pcpu_refcnt) - goto free_tx; + goto free_p; if (dev_addr_init(dev)) goto free_pcpu; @@ -5553,7 +5558,6 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, dev_net_set(dev, &init_net); - dev->_tx = tx; dev->num_tx_queues = queue_count; dev->real_num_tx_queues = queue_count; @@ -5564,8 +5568,6 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, dev->gso_max_size = GSO_MAX_SIZE; - netdev_init_queues(dev); - INIT_LIST_HEAD(&dev->ethtool_ntuple_list.list); dev->ethtool_ntuple_list.count = 0; INIT_LIST_HEAD(&dev->napi_list); @@ -5576,8 +5578,6 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, strcpy(dev->name, name); return dev; -free_tx: - kfree(tx); free_pcpu: free_percpu(dev->pcpu_refcnt); free_p: -- cgit v1.2.3 From 27b75c95f10d249574d9c4cb9dab878107faede8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 15 Oct 2010 05:44:11 +0000 Subject: net: avoid RCU for NOCACHE dst There is no point using RCU for dst we allocate for a very short time (used once). Change dst_release() to take DST_NOCACHE into account, but also change skb_dst_set_noref() to force a refcount increment for such dst. This is a _huge_ gain, because we dont waste memory to store xx thousand of dsts. Instead of queueing them to RCU, we can free them instantly. CPU caches can stay hot, re-using same memory blocks to hold temporary dsts. Note : remove unneeded smp_mb__before_atomic_dec(); in dst_release(), since atomic_dec_return() implies a full memory barrier. Stress test, 160.000.000 udp frames sent, IP route cache disabled (DDOS). Before: real 0m38.091s user 0m13.189s sys 7m53.018s After: real 0m29.946s user 0m12.157s sys 7m40.605s For reference, if IP route cache was enabled : real 0m32.030s user 0m10.521s sys 8m15.243s Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/skbuff.h | 14 +------------- net/core/dst.c | 29 ++++++++++++++++++++++++++++- net/ipv4/route.c | 9 ++++----- 3 files changed, 33 insertions(+), 19 deletions(-) (limited to 'net/core') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 05a358f1ba11..e6ba898de61c 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -460,19 +460,7 @@ static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst) skb->_skb_refdst = (unsigned long)dst; } -/** - * skb_dst_set_noref - sets skb dst, without a reference - * @skb: buffer - * @dst: dst entry - * - * Sets skb dst, assuming a reference was not taken on dst - * skb_dst_drop() should not dst_release() this dst - */ -static inline void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst) -{ - WARN_ON(!rcu_read_lock_held() && !rcu_read_lock_bh_held()); - skb->_skb_refdst = (unsigned long)dst | SKB_DST_NOREF; -} +extern void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst); /** * skb_dst_is_noref - Test if skb dst isnt refcounted diff --git a/net/core/dst.c b/net/core/dst.c index 32e542d7f472..8abe628b79f1 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -271,13 +271,40 @@ void dst_release(struct dst_entry *dst) if (dst) { int newrefcnt; - smp_mb__before_atomic_dec(); newrefcnt = atomic_dec_return(&dst->__refcnt); WARN_ON(newrefcnt < 0); + if (unlikely(dst->flags & DST_NOCACHE) && !newrefcnt) { + dst = dst_destroy(dst); + if (dst) + __dst_free(dst); + } } } EXPORT_SYMBOL(dst_release); +/** + * skb_dst_set_noref - sets skb dst, without a reference + * @skb: buffer + * @dst: dst entry + * + * Sets skb dst, assuming a reference was not taken on dst + * skb_dst_drop() should not dst_release() this dst + */ +void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst) +{ + WARN_ON(!rcu_read_lock_held() && !rcu_read_lock_bh_held()); + /* If dst not in cache, we must take a reference, because + * dst_release() will destroy dst as soon as its refcount becomes zero + */ + if (unlikely(dst->flags & DST_NOCACHE)) { + dst_hold(dst); + skb_dst_set(skb, dst); + } else { + skb->_skb_refdst = (unsigned long)dst | SKB_DST_NOREF; + } +} +EXPORT_SYMBOL(skb_dst_set_noref); + /* Dirty hack. We did it in 2.2 (in __dst_free), * we have _very_ good reasons not to repeat * this mistake in 2.3, but we have no choice diff --git a/net/ipv4/route.c b/net/ipv4/route.c index ff98983d2a45..d6cb2bfcd8e1 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1105,9 +1105,9 @@ restart: * Note that we do rt_free on this new route entry, so that * once its refcount hits zero, we are still able to reap it * (Thanks Alexey) - * Note also the rt_free uses call_rcu. We don't actually - * need rcu protection here, this is just our path to get - * on the route gc list. + * Note: To avoid expensive rcu stuff for this uncached dst, + * we set DST_NOCACHE so that dst_release() can free dst without + * waiting a grace period. */ rt->dst.flags |= DST_NOCACHE; @@ -1117,12 +1117,11 @@ restart: if (net_ratelimit()) printk(KERN_WARNING "Neighbour table failure & not caching routes.\n"); - rt_drop(rt); + ip_rt_put(rt); return err; } } - rt_free(rt); goto skip_hashing; } -- cgit v1.2.3 From 7b9c60903714bf0a19d746b228864bad3497284e Mon Sep 17 00:00:00 2001 From: Jesse Gross Date: Wed, 20 Oct 2010 13:56:04 +0000 Subject: vlan: Enable software emulation for vlan accleration. Currently users of hardware vlan accleration need to know whether the device supports it before generating packets. However, vlan acceleration will soon be available in a more flexible manner so knowing ahead of time becomes much more difficult. This adds a software fallback path for vlan packets on devices without the necessary offloading support, similar to other types of hardware accleration. Signed-off-by: Jesse Gross Signed-off-by: David S. Miller --- include/linux/netdevice.h | 14 +++++++++++--- net/core/dev.c | 36 +++++++++++++++++++++++++++++++++--- 2 files changed, 44 insertions(+), 6 deletions(-) (limited to 'net/core') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 880d56565828..2861565a27d9 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2248,9 +2248,17 @@ static inline int skb_gso_ok(struct sk_buff *skb, int features) static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb) { - return skb_is_gso(skb) && - (!skb_gso_ok(skb, dev->features) || - unlikely(skb->ip_summed != CHECKSUM_PARTIAL)); + if (skb_is_gso(skb)) { + int features = dev->features; + + if (skb->protocol == htons(ETH_P_8021Q) || skb->vlan_tci) + features &= dev->vlan_features; + + return (!skb_gso_ok(skb, features) || + unlikely(skb->ip_summed != CHECKSUM_PARTIAL)); + } + + return 0; } static inline void netif_set_gso_max_size(struct net_device *dev, diff --git a/net/core/dev.c b/net/core/dev.c index 4c3ac53e4b16..1bfd96b1fbd4 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1694,7 +1694,12 @@ static bool can_checksum_protocol(unsigned long features, __be16 protocol) static bool dev_can_checksum(struct net_device *dev, struct sk_buff *skb) { - if (can_checksum_protocol(dev->features, skb->protocol)) + int features = dev->features; + + if (vlan_tx_tag_present(skb)) + features &= dev->vlan_features; + + if (can_checksum_protocol(features, skb->protocol)) return true; if (skb->protocol == htons(ETH_P_8021Q)) { @@ -1793,6 +1798,16 @@ struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features) __be16 type = skb->protocol; int err; + if (type == htons(ETH_P_8021Q)) { + struct vlan_ethhdr *veh; + + if (unlikely(!pskb_may_pull(skb, VLAN_ETH_HLEN))) + return ERR_PTR(-EINVAL); + + veh = (struct vlan_ethhdr *)skb->data; + type = veh->h_vlan_encapsulated_proto; + } + skb_reset_mac_header(skb); skb->mac_len = skb->network_header - skb->mac_header; __skb_pull(skb, skb->mac_len); @@ -1964,9 +1979,14 @@ static inline void skb_orphan_try(struct sk_buff *skb) static inline int skb_needs_linearize(struct sk_buff *skb, struct net_device *dev) { + int features = dev->features; + + if (skb->protocol == htons(ETH_P_8021Q) || vlan_tx_tag_present(skb)) + features &= dev->vlan_features; + return skb_is_nonlinear(skb) && - ((skb_has_frag_list(skb) && !(dev->features & NETIF_F_FRAGLIST)) || - (skb_shinfo(skb)->nr_frags && (!(dev->features & NETIF_F_SG) || + ((skb_has_frag_list(skb) && !(features & NETIF_F_FRAGLIST)) || + (skb_shinfo(skb)->nr_frags && (!(features & NETIF_F_SG) || illegal_highdma(dev, skb)))); } @@ -1989,6 +2009,15 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, skb_orphan_try(skb); + if (vlan_tx_tag_present(skb) && + !(dev->features & NETIF_F_HW_VLAN_TX)) { + skb = __vlan_put_tag(skb, vlan_tx_tag_get(skb)); + if (unlikely(!skb)) + goto out; + + skb->vlan_tci = 0; + } + if (netif_needs_gso(dev, skb)) { if (unlikely(dev_gso_segment(skb))) goto out_kfree_skb; @@ -2050,6 +2079,7 @@ out_kfree_gso_skb: skb->destructor = DEV_GSO_CB(skb)->destructor; out_kfree_skb: kfree_skb(skb); +out: return rc; } -- cgit v1.2.3 From 3701e51382a026cba10c60b03efabe534fba4ca4 Mon Sep 17 00:00:00 2001 From: Jesse Gross Date: Wed, 20 Oct 2010 13:56:06 +0000 Subject: vlan: Centralize handling of hardware acceleration. Currently each driver that is capable of vlan hardware acceleration must be aware of the vlan groups that are configured and then pass the stripped tag to a specialized receive function. This is different from other types of hardware offload in that it places a significant amount of knowledge in the driver itself rather keeping it in the networking core. This makes vlan offloading function more similarly to other forms of offloading (such as checksum offloading or TSO) by doing the following: * On receive, stripped vlans are passed directly to the network core, without attempting to check for vlan groups or reconstructing the header if no group * vlans are made less special by folding the logic into the main receive routines * On transmit, the device layer will add the vlan header in software if the hardware doesn't support it, instead of spreading that logic out in upper layers, such as bonding. There are a number of advantages to this: * Fixes all bugs with drivers incorrectly dropping vlan headers at once. * Avoids having to disable VLAN acceleration when in promiscuous mode (good for bridging since it always puts devices in promiscuous mode). * Keeps VLAN tag separate until given to ultimate consumer, which avoids needing to do header reconstruction as in tg3 unless absolutely necessary. * Consolidates common code in core networking. Signed-off-by: Jesse Gross Signed-off-by: David S. Miller --- include/linux/if_vlan.h | 6 ++- include/linux/netdevice.h | 1 - net/8021q/vlan.c | 9 +--- net/8021q/vlan_core.c | 125 ++++++++++------------------------------------ net/core/dev.c | 47 ++++++----------- 5 files changed, 48 insertions(+), 140 deletions(-) (limited to 'net/core') diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 4047781da727..a0d9786c202d 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -132,7 +132,7 @@ extern u16 vlan_dev_vlan_id(const struct net_device *dev); extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, u16 vlan_tci, int polling); -extern void vlan_hwaccel_do_receive(struct sk_buff *skb); +extern bool vlan_hwaccel_do_receive(struct sk_buff **skb); extern gro_result_t vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp, unsigned int vlan_tci, struct sk_buff *skb); @@ -166,8 +166,10 @@ static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, return NET_XMIT_SUCCESS; } -static inline void vlan_hwaccel_do_receive(struct sk_buff *skb) +static inline bool vlan_hwaccel_do_receive(struct sk_buff **skb) { + BUG(); + return false; } static inline gro_result_t diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9c78312ce142..ed7db7eebbf3 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1768,7 +1768,6 @@ extern int netdev_rx_handler_register(struct net_device *dev, void *rx_handler_data); extern void netdev_rx_handler_unregister(struct net_device *dev); -extern void netif_nit_deliver(struct sk_buff *skb); extern int dev_valid_name(const char *name); extern int dev_ioctl(struct net *net, unsigned int cmd, void __user *); extern int dev_ethtool(struct net *net, struct ifreq *); diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index f862dccf6bb0..05b867e43757 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -135,7 +135,7 @@ void unregister_vlan_dev(struct net_device *dev, struct list_head *head) vlan_gvrp_uninit_applicant(real_dev); rcu_assign_pointer(real_dev->vlgrp, NULL); - if (real_dev->features & NETIF_F_HW_VLAN_RX) + if (ops->ndo_vlan_rx_register) ops->ndo_vlan_rx_register(real_dev, NULL); /* Free the group, after all cpu's are done. */ @@ -156,11 +156,6 @@ int vlan_check_real_dev(struct net_device *real_dev, u16 vlan_id) return -EOPNOTSUPP; } - if ((real_dev->features & NETIF_F_HW_VLAN_RX) && !ops->ndo_vlan_rx_register) { - pr_info("8021q: device %s has buggy VLAN hw accel\n", name); - return -EOPNOTSUPP; - } - if ((real_dev->features & NETIF_F_HW_VLAN_FILTER) && (!ops->ndo_vlan_rx_add_vid || !ops->ndo_vlan_rx_kill_vid)) { pr_info("8021q: Device %s has buggy VLAN hw accel\n", name); @@ -213,7 +208,7 @@ int register_vlan_dev(struct net_device *dev) grp->nr_vlans++; if (ngrp) { - if (real_dev->features & NETIF_F_HW_VLAN_RX) + if (ops->ndo_vlan_rx_register) ops->ndo_vlan_rx_register(real_dev, ngrp); rcu_assign_pointer(real_dev->vlgrp, ngrp); } diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index dee727ce0291..69b2f79800a5 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -4,54 +4,29 @@ #include #include "vlan.h" -/* VLAN rx hw acceleration helper. This acts like netif_{rx,receive_skb}(). */ -int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, - u16 vlan_tci, int polling) +bool vlan_hwaccel_do_receive(struct sk_buff **skbp) { + struct sk_buff *skb = *skbp; + u16 vlan_id = skb->vlan_tci & VLAN_VID_MASK; struct net_device *vlan_dev; - u16 vlan_id; - - if (netpoll_rx(skb)) - return NET_RX_DROP; - - if (skb_bond_should_drop(skb, ACCESS_ONCE(skb->dev->master))) - skb->deliver_no_wcard = 1; + struct vlan_rx_stats *rx_stats; - skb->skb_iif = skb->dev->ifindex; - __vlan_hwaccel_put_tag(skb, vlan_tci); - vlan_id = vlan_tci & VLAN_VID_MASK; - vlan_dev = vlan_group_get_device(grp, vlan_id); - - if (vlan_dev) - skb->dev = vlan_dev; - else if (vlan_id) { - if (!(skb->dev->flags & IFF_PROMISC)) - goto drop; - skb->pkt_type = PACKET_OTHERHOST; + vlan_dev = vlan_find_dev(skb->dev, vlan_id); + if (!vlan_dev) { + if (vlan_id) + skb->pkt_type = PACKET_OTHERHOST; + return false; } - return polling ? netif_receive_skb(skb) : netif_rx(skb); + skb = *skbp = skb_share_check(skb, GFP_ATOMIC); + if (unlikely(!skb)) + return false; -drop: - atomic_long_inc(&skb->dev->rx_dropped); - dev_kfree_skb_any(skb); - return NET_RX_DROP; -} -EXPORT_SYMBOL(__vlan_hwaccel_rx); - -void vlan_hwaccel_do_receive(struct sk_buff *skb) -{ - struct net_device *dev = skb->dev; - struct vlan_rx_stats *rx_stats; - - skb->dev = vlan_dev_real_dev(dev); - netif_nit_deliver(skb); - - skb->dev = dev; - skb->priority = vlan_get_ingress_priority(dev, skb->vlan_tci); + skb->dev = vlan_dev; + skb->priority = vlan_get_ingress_priority(vlan_dev, skb->vlan_tci); skb->vlan_tci = 0; - rx_stats = this_cpu_ptr(vlan_dev_info(dev)->vlan_rx_stats); + rx_stats = this_cpu_ptr(vlan_dev_info(vlan_dev)->vlan_rx_stats); u64_stats_update_begin(&rx_stats->syncp); rx_stats->rx_packets++; @@ -68,11 +43,13 @@ void vlan_hwaccel_do_receive(struct sk_buff *skb) * This allows the VLAN to have a different MAC than the * underlying device, and still route correctly. */ if (!compare_ether_addr(eth_hdr(skb)->h_dest, - dev->dev_addr)) + vlan_dev->dev_addr)) skb->pkt_type = PACKET_HOST; break; } u64_stats_update_end(&rx_stats->syncp); + + return true; } struct net_device *vlan_dev_real_dev(const struct net_device *dev) @@ -87,75 +64,27 @@ u16 vlan_dev_vlan_id(const struct net_device *dev) } EXPORT_SYMBOL(vlan_dev_vlan_id); -static gro_result_t -vlan_gro_common(struct napi_struct *napi, struct vlan_group *grp, - unsigned int vlan_tci, struct sk_buff *skb) +/* VLAN rx hw acceleration helper. This acts like netif_{rx,receive_skb}(). */ +int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, + u16 vlan_tci, int polling) { - struct sk_buff *p; - struct net_device *vlan_dev; - u16 vlan_id; - - if (skb_bond_should_drop(skb, ACCESS_ONCE(skb->dev->master))) - skb->deliver_no_wcard = 1; - - skb->skb_iif = skb->dev->ifindex; __vlan_hwaccel_put_tag(skb, vlan_tci); - vlan_id = vlan_tci & VLAN_VID_MASK; - vlan_dev = vlan_group_get_device(grp, vlan_id); - - if (vlan_dev) - skb->dev = vlan_dev; - else if (vlan_id) { - if (!(skb->dev->flags & IFF_PROMISC)) - goto drop; - skb->pkt_type = PACKET_OTHERHOST; - } - - for (p = napi->gro_list; p; p = p->next) { - unsigned long diffs; - - diffs = (unsigned long)p->dev ^ (unsigned long)skb->dev; - diffs |= compare_ether_header(skb_mac_header(p), - skb_gro_mac_header(skb)); - NAPI_GRO_CB(p)->same_flow = !diffs; - NAPI_GRO_CB(p)->flush = 0; - } - - return dev_gro_receive(napi, skb); - -drop: - atomic_long_inc(&skb->dev->rx_dropped); - return GRO_DROP; + return polling ? netif_receive_skb(skb) : netif_rx(skb); } +EXPORT_SYMBOL(__vlan_hwaccel_rx); gro_result_t vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp, unsigned int vlan_tci, struct sk_buff *skb) { - if (netpoll_rx_on(skb)) - return vlan_hwaccel_receive_skb(skb, grp, vlan_tci) - ? GRO_DROP : GRO_NORMAL; - - skb_gro_reset_offset(skb); - - return napi_skb_finish(vlan_gro_common(napi, grp, vlan_tci, skb), skb); + __vlan_hwaccel_put_tag(skb, vlan_tci); + return napi_gro_receive(napi, skb); } EXPORT_SYMBOL(vlan_gro_receive); gro_result_t vlan_gro_frags(struct napi_struct *napi, struct vlan_group *grp, unsigned int vlan_tci) { - struct sk_buff *skb = napi_frags_skb(napi); - - if (!skb) - return GRO_DROP; - - if (netpoll_rx_on(skb)) { - skb->protocol = eth_type_trans(skb, skb->dev); - return vlan_hwaccel_receive_skb(skb, grp, vlan_tci) - ? GRO_DROP : GRO_NORMAL; - } - - return napi_frags_finish(napi, skb, - vlan_gro_common(napi, grp, vlan_tci, skb)); + __vlan_hwaccel_put_tag(napi->skb, vlan_tci); + return napi_gro_frags(napi); } EXPORT_SYMBOL(vlan_gro_frags); diff --git a/net/core/dev.c b/net/core/dev.c index 1bfd96b1fbd4..97fd6bc2004c 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2789,33 +2789,6 @@ out: } #endif -/* - * netif_nit_deliver - deliver received packets to network taps - * @skb: buffer - * - * This function is used to deliver incoming packets to network - * taps. It should be used when the normal netif_receive_skb path - * is bypassed, for example because of VLAN acceleration. - */ -void netif_nit_deliver(struct sk_buff *skb) -{ - struct packet_type *ptype; - - if (list_empty(&ptype_all)) - return; - - skb_reset_network_header(skb); - skb_reset_transport_header(skb); - skb->mac_len = skb->network_header - skb->mac_header; - - rcu_read_lock(); - list_for_each_entry_rcu(ptype, &ptype_all, list) { - if (!ptype->dev || ptype->dev == skb->dev) - deliver_skb(skb, ptype, skb->dev); - } - rcu_read_unlock(); -} - /** * netdev_rx_handler_register - register receive handler * @dev: device to register a handler for @@ -2925,9 +2898,6 @@ static int __netif_receive_skb(struct sk_buff *skb) if (!netdev_tstamp_prequeue) net_timestamp_check(skb); - if (vlan_tx_tag_present(skb)) - vlan_hwaccel_do_receive(skb); - /* if we've gotten here through NAPI, check netpoll */ if (netpoll_receive_skb(skb)) return NET_RX_DROP; @@ -2940,8 +2910,7 @@ static int __netif_receive_skb(struct sk_buff *skb) * be delivered to pkt handlers that are exact matches. Also * the deliver_no_wcard flag will be set. If packet handlers * are sensitive to duplicate packets these skbs will need to - * be dropped at the handler. The vlan accel path may have - * already set the deliver_no_wcard flag. + * be dropped at the handler. */ null_or_orig = NULL; orig_dev = skb->dev; @@ -3000,6 +2969,18 @@ ncls: goto out; } + if (vlan_tx_tag_present(skb)) { + if (pt_prev) { + ret = deliver_skb(skb, pt_prev, orig_dev); + pt_prev = NULL; + } + if (vlan_hwaccel_do_receive(&skb)) { + ret = __netif_receive_skb(skb); + goto out; + } else if (unlikely(!skb)) + goto out; + } + /* * Make sure frames received on VLAN interfaces stacked on * bonding interfaces still make their way to any base bonding @@ -3264,6 +3245,7 @@ __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) unsigned long diffs; diffs = (unsigned long)p->dev ^ (unsigned long)skb->dev; + diffs |= p->vlan_tci ^ skb->vlan_tci; diffs |= compare_ether_header(skb_mac_header(p), skb_gro_mac_header(skb)); NAPI_GRO_CB(p)->same_flow = !diffs; @@ -3323,6 +3305,7 @@ void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) { __skb_pull(skb, skb_headlen(skb)); skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb)); + skb->vlan_tci = 0; napi->skb = skb; } -- cgit v1.2.3 From d5dbda23804156ae6f35025ade5307a49d1db6d7 Mon Sep 17 00:00:00 2001 From: Jesse Gross Date: Wed, 20 Oct 2010 13:56:07 +0000 Subject: ethtool: Add support for vlan accleration. Now that vlan acceleration is handled consistently regardless of usage, it is possible to enable and disable it at will. This adds support for Ethtool operations that change the offloading status for debugging purposes, similar to other forms of hardware acceleration. Signed-off-by: Jesse Gross Signed-off-by: David S. Miller --- include/linux/ethtool.h | 2 ++ net/core/ethtool.c | 3 ++- 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'net/core') diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index 8a3338ceb438..6628a507fd3b 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -309,6 +309,8 @@ struct ethtool_perm_addr { * flag differs from the read-only value. */ enum ethtool_flags { + ETH_FLAG_TXVLAN = (1 << 7), /* TX VLAN offload enabled */ + ETH_FLAG_RXVLAN = (1 << 8), /* RX VLAN offload enabled */ ETH_FLAG_LRO = (1 << 15), /* LRO is enabled */ ETH_FLAG_NTUPLE = (1 << 27), /* N-tuple filters enabled */ ETH_FLAG_RXHASH = (1 << 28), diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 685c7005e87f..956a9f4971cb 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -132,7 +132,8 @@ EXPORT_SYMBOL(ethtool_op_set_ufo); * NETIF_F_xxx values in include/linux/netdevice.h */ static const u32 flags_dup_features = - (ETH_FLAG_LRO | ETH_FLAG_NTUPLE | ETH_FLAG_RXHASH); + (ETH_FLAG_LRO | ETH_FLAG_RXVLAN | ETH_FLAG_TXVLAN | ETH_FLAG_NTUPLE | + ETH_FLAG_RXHASH); u32 ethtool_op_get_flags(struct net_device *dev) { -- cgit v1.2.3 From 8d8a0b1cc2a8f9794a3f1f747089b6a93774408d Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Fri, 15 Oct 2010 05:12:01 +0000 Subject: rtnetlink: remove rtnl_kill_links The function rtnl_kill_links is defined but never used. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/net/rtnetlink.h | 1 - net/core/rtnetlink.c | 8 -------- 2 files changed, 9 deletions(-) (limited to 'net/core') diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index af60fd050844..e013c68bfb00 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -79,7 +79,6 @@ struct rtnl_link_ops { extern int __rtnl_link_register(struct rtnl_link_ops *ops); extern void __rtnl_link_unregister(struct rtnl_link_ops *ops); -extern void rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops); extern int rtnl_link_register(struct rtnl_link_ops *ops); extern void rtnl_link_unregister(struct rtnl_link_ops *ops); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index b2a718dfd720..8121268ddbdd 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -299,14 +299,6 @@ static void __rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops) unregister_netdevice_many(&list_kill); } -void rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops) -{ - rtnl_lock(); - __rtnl_kill_links(net, ops); - rtnl_unlock(); -} -EXPORT_SYMBOL_GPL(rtnl_kill_links); - /** * __rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink. * @ops: struct rtnl_link_ops * to unregister -- cgit v1.2.3 From d2ed817766987fd05e69b7da65d4861b38f1aa2a Mon Sep 17 00:00:00 2001 From: Ben Greear Date: Thu, 21 Oct 2010 04:06:29 -0700 Subject: net/core: Allow tagged VLAN packets to flow through VETH devices. When there are VLANs on a VETH device, the packets being transmitted through the VETH device may be 4 bytes bigger than MTU. A check in dev_forward_skb did not take this into account and so dropped these packets. This patch is needed at least as far back as 2.6.34.7 and should be considered for -stable. Signed-off-by: Ben Greear Signed-off-by: David S. Miller --- net/core/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 660dd41aaaa6..8e07109cc0ef 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1485,7 +1485,7 @@ int dev_forward_skb(struct net_device *dev, struct sk_buff *skb) nf_reset(skb); if (!(dev->flags & IFF_UP) || - (skb->len > (dev->mtu + dev->hard_header_len))) { + (skb->len > (dev->mtu + dev->hard_header_len + VLAN_HLEN))) { kfree_skb(skb); return NET_RX_DROP; } -- cgit v1.2.3 From a5c30b349b872aa2ac13babbd5ceb26737f17e95 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 19 Oct 2010 06:04:42 +0000 Subject: net/neighbour: cancel_delayed_work() + flush_scheduled_work() -> cancel_delayed_work_sync() flush_scheduled_work() is going away. Prepare for it. Signed-off-by: Tejun Heo Signed-off-by: David S. Miller --- net/core/neighbour.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net/core') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index b165b96355bf..8cc8f9a79db9 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1588,8 +1588,7 @@ int neigh_table_clear(struct neigh_table *tbl) struct neigh_table **tp; /* It is not clean... Fix it to unload IPv6 module safely */ - cancel_delayed_work(&tbl->gc_work); - flush_scheduled_work(); + cancel_delayed_work_sync(&tbl->gc_work); del_timer_sync(&tbl->proxy_timer); pneigh_queue_purge(&tbl->proxy_queue); neigh_ifdown(tbl, NULL); -- cgit v1.2.3 From d0c2b0d265a0f1f92922a99a31def9da582197ac Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Tue, 19 Oct 2010 07:12:10 +0000 Subject: napi: unexport napi_reuse_skb The function napi_reuse_skb is only used inside core. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 -- net/core/dev.c | 3 +-- 2 files changed, 1 insertion(+), 4 deletions(-) (limited to 'net/core') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ed7db7eebbf3..fcd3dda86322 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1748,8 +1748,6 @@ extern gro_result_t napi_skb_finish(gro_result_t ret, struct sk_buff *skb); extern gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb); extern void napi_gro_flush(struct napi_struct *napi); -extern void napi_reuse_skb(struct napi_struct *napi, - struct sk_buff *skb); extern struct sk_buff * napi_get_frags(struct napi_struct *napi); extern gro_result_t napi_frags_finish(struct napi_struct *napi, struct sk_buff *skb, diff --git a/net/core/dev.c b/net/core/dev.c index 97fd6bc2004c..b2269ac04cb8 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3301,7 +3301,7 @@ gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) } EXPORT_SYMBOL(napi_gro_receive); -void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) +static void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) { __skb_pull(skb, skb_headlen(skb)); skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb)); @@ -3309,7 +3309,6 @@ void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) napi->skb = skb; } -EXPORT_SYMBOL(napi_reuse_skb); struct sk_buff *napi_get_frags(struct napi_struct *napi) { -- cgit v1.2.3