提交 4e588793 编写于 作者: D David S. Miller

Merge branch 'neigh'

Jiri Pirko says:

====================
neigh: respect default parms values

This is a long standing regression. But since the patchset is bigger and
the regression happened in 2007, I'm proposing this to net-next instead.

Basically the problem is that if user wants to use /etc/sysctl.conf to specify
default values of neigh related params, he is not able to do that.

The reason is that the default values are copied to dev instance right after
netdev is registered. And that is way to early. The original behaviour
for ipv4 was that this happened after first address was assigned to device.
For ipv6 this was apparently from the very beginning.

So this patchset basically reverts the behaviour back to what it was in 2007 for
ipv4 and changes the behaviour for ipv6 so they are both the same.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
......@@ -220,6 +220,13 @@ static inline struct in_device *__in_dev_get_rtnl(const struct net_device *dev)
return rtnl_dereference(dev->ip_ptr);
}
static inline struct neigh_parms *__in_dev_arp_parms_get_rcu(const struct net_device *dev)
{
struct in_device *in_dev = __in_dev_get_rcu(dev);
return in_dev ? in_dev->arp_parms : NULL;
}
void in_dev_finish_destroy(struct in_device *idev);
static inline void in_dev_put(struct in_device *idev)
......
......@@ -248,6 +248,13 @@ static inline struct inet6_dev *in6_dev_get(const struct net_device *dev)
return idev;
}
static inline struct neigh_parms *__in6_dev_nd_parms_get_rcu(const struct net_device *dev)
{
struct inet6_dev *idev = __in6_dev_get(dev);
return idev ? idev->nd_parms : NULL;
}
void in6_dev_finish_destroy(struct inet6_dev *idev);
static inline void in6_dev_put(struct inet6_dev *idev)
......
......@@ -21,6 +21,7 @@
#include <linux/skbuff.h>
#include <linux/rcupdate.h>
#include <linux/seq_file.h>
#include <linux/bitmap.h>
#include <linux/err.h>
#include <linux/sysctl.h>
......@@ -37,6 +38,32 @@
struct neighbour;
enum {
NEIGH_VAR_MCAST_PROBES,
NEIGH_VAR_UCAST_PROBES,
NEIGH_VAR_APP_PROBES,
NEIGH_VAR_RETRANS_TIME,
NEIGH_VAR_BASE_REACHABLE_TIME,
NEIGH_VAR_DELAY_PROBE_TIME,
NEIGH_VAR_GC_STALETIME,
NEIGH_VAR_QUEUE_LEN_BYTES,
NEIGH_VAR_PROXY_QLEN,
NEIGH_VAR_ANYCAST_DELAY,
NEIGH_VAR_PROXY_DELAY,
NEIGH_VAR_LOCKTIME,
#define NEIGH_VAR_DATA_MAX (NEIGH_VAR_LOCKTIME + 1)
/* Following are used as a second way to access one of the above */
NEIGH_VAR_QUEUE_LEN, /* same data as NEIGH_VAR_QUEUE_LEN_BYTES */
NEIGH_VAR_RETRANS_TIME_MS, /* same data as NEIGH_VAR_RETRANS_TIME */
NEIGH_VAR_BASE_REACHABLE_TIME_MS, /* same data as NEIGH_VAR_BASE_REACHABLE_TIME */
/* Following are used by "default" only */
NEIGH_VAR_GC_INTERVAL,
NEIGH_VAR_GC_THRESH1,
NEIGH_VAR_GC_THRESH2,
NEIGH_VAR_GC_THRESH3,
NEIGH_VAR_MAX
};
struct neigh_parms {
#ifdef CONFIG_NET_NS
struct net *net;
......@@ -53,22 +80,30 @@ struct neigh_parms {
atomic_t refcnt;
struct rcu_head rcu_head;
int base_reachable_time;
int retrans_time;
int gc_staletime;
int reachable_time;
int delay_probe_time;
int queue_len_bytes;
int ucast_probes;
int app_probes;
int mcast_probes;
int anycast_delay;
int proxy_delay;
int proxy_qlen;
int locktime;
int data[NEIGH_VAR_DATA_MAX];
DECLARE_BITMAP(data_state, NEIGH_VAR_DATA_MAX);
};
static inline void neigh_var_set(struct neigh_parms *p, int index, int val)
{
set_bit(index, p->data_state);
p->data[index] = val;
}
#define NEIGH_VAR(p, attr) ((p)->data[NEIGH_VAR_ ## attr])
#define NEIGH_VAR_SET(p, attr, val) neigh_var_set(p, NEIGH_VAR_ ## attr, val)
static inline void neigh_parms_data_state_setall(struct neigh_parms *p)
{
bitmap_fill(p->data_state, NEIGH_VAR_DATA_MAX);
}
static inline void neigh_parms_data_state_cleanall(struct neigh_parms *p)
{
bitmap_zero(p->data_state, NEIGH_VAR_DATA_MAX);
}
struct neigh_statistics {
unsigned long allocs; /* number of allocated neighs */
unsigned long destroys; /* number of destroyed neighs */
......@@ -180,6 +215,11 @@ struct neigh_table {
struct pneigh_entry **phash_buckets;
};
static inline int neigh_parms_family(struct neigh_parms *p)
{
return p->tbl->family;
}
#define NEIGH_PRIV_ALIGN sizeof(long long)
#define NEIGH_ENTRY_SIZE(size) ALIGN((size), NEIGH_PRIV_ALIGN)
......@@ -274,8 +314,17 @@ void *neigh_seq_start(struct seq_file *, loff_t *, struct neigh_table *,
void *neigh_seq_next(struct seq_file *, void *, loff_t *);
void neigh_seq_stop(struct seq_file *, void *);
int neigh_proc_dointvec(struct ctl_table *ctl, int write,
void __user *buffer, size_t *lenp, loff_t *ppos);
int neigh_proc_dointvec_jiffies(struct ctl_table *ctl, int write,
void __user *buffer,
size_t *lenp, loff_t *ppos);
int neigh_proc_dointvec_ms_jiffies(struct ctl_table *ctl, int write,
void __user *buffer,
size_t *lenp, loff_t *ppos);
int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p,
char *p_name, proc_handler *proc_handler);
proc_handler *proc_handler);
void neigh_sysctl_unregister(struct neigh_parms *p);
static inline void __neigh_parms_put(struct neigh_parms *parms)
......
......@@ -172,14 +172,14 @@ EXPORT_SYMBOL(hippi_mac_addr);
int hippi_neigh_setup_dev(struct net_device *dev, struct neigh_parms *p)
{
/* Never send broadcast/multicast ARP messages */
p->mcast_probes = 0;
NEIGH_VAR_SET(p, MCAST_PROBES, 0);
/* In IPv6 unicast probes are valid even on NBMA,
* because they are encapsulated in normal IPv6 protocol.
* Should be a generic flag.
*/
if (p->tbl->family != AF_INET6)
p->ucast_probes = 0;
NEIGH_VAR_SET(p, UCAST_PROBES, 0);
return 0;
}
EXPORT_SYMBOL(hippi_neigh_setup_dev);
......
此差异已折叠。
......@@ -102,19 +102,21 @@ struct neigh_table dn_neigh_table = {
.id = "dn_neigh_cache",
.parms ={
.tbl = &dn_neigh_table,
.base_reachable_time = 30 * HZ,
.retrans_time = 1 * HZ,
.gc_staletime = 60 * HZ,
.reachable_time = 30 * HZ,
.delay_probe_time = 5 * HZ,
.queue_len_bytes = 64*1024,
.ucast_probes = 0,
.app_probes = 0,
.mcast_probes = 0,
.anycast_delay = 0,
.proxy_delay = 0,
.proxy_qlen = 0,
.locktime = 1 * HZ,
.reachable_time = 30 * HZ,
.data = {
[NEIGH_VAR_MCAST_PROBES] = 0,
[NEIGH_VAR_UCAST_PROBES] = 0,
[NEIGH_VAR_APP_PROBES] = 0,
[NEIGH_VAR_RETRANS_TIME] = 1 * HZ,
[NEIGH_VAR_BASE_REACHABLE_TIME] = 30 * HZ,
[NEIGH_VAR_DELAY_PROBE_TIME] = 5 * HZ,
[NEIGH_VAR_GC_STALETIME] = 60 * HZ,
[NEIGH_VAR_QUEUE_LEN_BYTES] = 64*1024,
[NEIGH_VAR_PROXY_QLEN] = 0,
[NEIGH_VAR_ANYCAST_DELAY] = 0,
[NEIGH_VAR_PROXY_DELAY] = 0,
[NEIGH_VAR_LOCKTIME] = 1 * HZ,
},
},
.gc_interval = 30 * HZ,
.gc_thresh1 = 128,
......
......@@ -166,18 +166,20 @@ struct neigh_table arp_tbl = {
.id = "arp_cache",
.parms = {
.tbl = &arp_tbl,
.base_reachable_time = 30 * HZ,
.retrans_time = 1 * HZ,
.gc_staletime = 60 * HZ,
.reachable_time = 30 * HZ,
.delay_probe_time = 5 * HZ,
.queue_len_bytes = 64*1024,
.ucast_probes = 3,
.mcast_probes = 3,
.anycast_delay = 1 * HZ,
.proxy_delay = (8 * HZ) / 10,
.proxy_qlen = 64,
.locktime = 1 * HZ,
.data = {
[NEIGH_VAR_MCAST_PROBES] = 3,
[NEIGH_VAR_UCAST_PROBES] = 3,
[NEIGH_VAR_RETRANS_TIME] = 1 * HZ,
[NEIGH_VAR_BASE_REACHABLE_TIME] = 30 * HZ,
[NEIGH_VAR_DELAY_PROBE_TIME] = 5 * HZ,
[NEIGH_VAR_GC_STALETIME] = 60 * HZ,
[NEIGH_VAR_QUEUE_LEN_BYTES] = 64 * 1024,
[NEIGH_VAR_PROXY_QLEN] = 64,
[NEIGH_VAR_ANYCAST_DELAY] = 1 * HZ,
[NEIGH_VAR_PROXY_DELAY] = (8 * HZ) / 10,
[NEIGH_VAR_LOCKTIME] = 1 * HZ,
},
},
.gc_interval = 30 * HZ,
.gc_thresh1 = 128,
......@@ -359,14 +361,14 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb)
if (!saddr)
saddr = inet_select_addr(dev, target, RT_SCOPE_LINK);
probes -= neigh->parms->ucast_probes;
probes -= NEIGH_VAR(neigh->parms, UCAST_PROBES);
if (probes < 0) {
if (!(neigh->nud_state & NUD_VALID))
pr_debug("trying to ucast probe in NUD_INVALID\n");
neigh_ha_snapshot(dst_ha, neigh, dev);
dst_hw = dst_ha;
} else {
probes -= neigh->parms->app_probes;
probes -= NEIGH_VAR(neigh->parms, APP_PROBES);
if (probes < 0) {
neigh_app_ns(neigh);
return;
......@@ -871,7 +873,7 @@ static int arp_process(struct sk_buff *skb)
if (NEIGH_CB(skb)->flags & LOCALLY_ENQUEUED ||
skb->pkt_type == PACKET_HOST ||
in_dev->arp_parms->proxy_delay == 0) {
NEIGH_VAR(in_dev->arp_parms, PROXY_DELAY) == 0) {
arp_send(ARPOP_REPLY, ETH_P_ARP, sip,
dev, tip, sha, dev->dev_addr,
sha);
......@@ -910,7 +912,8 @@ static int arp_process(struct sk_buff *skb)
agents are active. Taking the first reply prevents
arp trashing and chooses the fastest router.
*/
override = time_after(jiffies, n->updated + n->parms->locktime);
override = time_after(jiffies, n->updated +
NEIGH_VAR(n->parms, LOCKTIME));
/* Broadcast replies and request packets
do not assert neighbour reachability.
......@@ -1284,7 +1287,7 @@ void __init arp_init(void)
dev_add_pack(&arp_packet_type);
arp_proc_init();
#ifdef CONFIG_SYSCTL
neigh_sysctl_register(NULL, &arp_tbl.parms, "ipv4", NULL);
neigh_sysctl_register(NULL, &arp_tbl.parms, NULL);
#endif
register_netdevice_notifier(&arp_netdev_notifier);
}
......
......@@ -500,6 +500,7 @@ static int inet_set_ifa(struct net_device *dev, struct in_ifaddr *ifa)
return -ENOBUFS;
}
ipv4_devconf_setall(in_dev);
neigh_parms_data_state_setall(in_dev->arp_parms);
if (ifa->ifa_dev != in_dev) {
WARN_ON(ifa->ifa_dev);
in_dev_hold(in_dev);
......@@ -747,6 +748,7 @@ static struct in_ifaddr *rtm_to_ifaddr(struct net *net, struct nlmsghdr *nlh,
goto errout;
ipv4_devconf_setall(in_dev);
neigh_parms_data_state_setall(in_dev->arp_parms);
in_dev_hold(in_dev);
if (tb[IFA_ADDRESS] == NULL)
......@@ -2160,7 +2162,7 @@ static void __devinet_sysctl_unregister(struct ipv4_devconf *cnf)
static void devinet_sysctl_register(struct in_device *idev)
{
neigh_sysctl_register(idev->dev, idev->arp_parms, "ipv4", NULL);
neigh_sysctl_register(idev->dev, idev->arp_parms, NULL);
__devinet_sysctl_register(dev_net(idev->dev), idev->dev->name,
&idev->cnf);
}
......
......@@ -425,6 +425,7 @@ struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v)
goto failure;
ipv4_devconf_setall(in_dev);
neigh_parms_data_state_setall(in_dev->arp_parms);
IPV4_DEVCONF(in_dev->cnf, RP_FILTER) = 0;
if (dev_open(dev))
......@@ -517,6 +518,7 @@ static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt)
}
ipv4_devconf_setall(in_dev);
neigh_parms_data_state_setall(in_dev->arp_parms);
IPV4_DEVCONF(in_dev->cnf, RP_FILTER) = 0;
rcu_read_unlock();
......
......@@ -834,6 +834,8 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr,
goto out;
}
neigh_parms_data_state_setall(idev->nd_parms);
ifa->addr = *addr;
if (peer_addr)
ifa->peer_addr = *peer_addr;
......@@ -1071,7 +1073,7 @@ static int ipv6_create_tempaddr(struct inet6_ifaddr *ifp, struct inet6_ifaddr *i
regen_advance = idev->cnf.regen_max_retry *
idev->cnf.dad_transmits *
idev->nd_parms->retrans_time / HZ;
NEIGH_VAR(idev->nd_parms, RETRANS_TIME) / HZ;
write_unlock_bh(&idev->lock);
/* A temporary address is created only if this calculated Preferred
......@@ -1888,7 +1890,8 @@ static void ipv6_regen_rndid(unsigned long data)
expires = jiffies +
idev->cnf.temp_prefered_lft * HZ -
idev->cnf.regen_max_retry * idev->cnf.dad_transmits * idev->nd_parms->retrans_time -
idev->cnf.regen_max_retry * idev->cnf.dad_transmits *
NEIGH_VAR(idev->nd_parms, RETRANS_TIME) -
idev->cnf.max_desync_factor * HZ;
if (time_before(expires, jiffies)) {
pr_warn("%s: too short regeneration interval; timer disabled for %s\n",
......@@ -3188,7 +3191,8 @@ static void addrconf_dad_timer(unsigned long data)
}
ifp->dad_probes--;
addrconf_mod_dad_timer(ifp, ifp->idev->nd_parms->retrans_time);
addrconf_mod_dad_timer(ifp,
NEIGH_VAR(ifp->idev->nd_parms, RETRANS_TIME));
spin_unlock(&ifp->lock);
write_unlock(&idev->lock);
......@@ -3509,7 +3513,7 @@ static void addrconf_verify(unsigned long foo)
!(ifp->flags&IFA_F_TENTATIVE)) {
unsigned long regen_advance = ifp->idev->cnf.regen_max_retry *
ifp->idev->cnf.dad_transmits *
ifp->idev->nd_parms->retrans_time / HZ;
NEIGH_VAR(ifp->idev->nd_parms, RETRANS_TIME) / HZ;
if (age >= ifp->prefered_lft - regen_advance) {
struct inet6_ifaddr *ifpub = ifp->ifpub;
......@@ -4231,7 +4235,7 @@ static int inet6_fill_ifla6_attrs(struct sk_buff *skb, struct inet6_dev *idev)
ci.max_reasm_len = IPV6_MAXPLEN;
ci.tstamp = cstamp_delta(idev->tstamp);
ci.reachable_time = jiffies_to_msecs(idev->nd_parms->reachable_time);
ci.retrans_time = jiffies_to_msecs(idev->nd_parms->retrans_time);
ci.retrans_time = jiffies_to_msecs(NEIGH_VAR(idev->nd_parms, RETRANS_TIME));
if (nla_put(skb, IFLA_INET6_CACHEINFO, sizeof(ci), &ci))
goto nla_put_failure;
nla = nla_reserve(skb, IFLA_INET6_CONF, DEVCONF_MAX * sizeof(s32));
......@@ -5026,7 +5030,7 @@ static void __addrconf_sysctl_unregister(struct ipv6_devconf *p)
static void addrconf_sysctl_register(struct inet6_dev *idev)
{
neigh_sysctl_register(idev->dev, idev->nd_parms, "ipv6",
neigh_sysctl_register(idev->dev, idev->nd_parms,
&ndisc_ifinfo_sysctl_change);
__addrconf_sysctl_register(dev_net(idev->dev), idev->dev->name,
idev, &idev->cnf);
......
......@@ -125,17 +125,19 @@ struct neigh_table nd_tbl = {
.id = "ndisc_cache",
.parms = {
.tbl = &nd_tbl,
.base_reachable_time = ND_REACHABLE_TIME,
.retrans_time = ND_RETRANS_TIMER,
.gc_staletime = 60 * HZ,
.reachable_time = ND_REACHABLE_TIME,
.delay_probe_time = 5 * HZ,
.queue_len_bytes = 64*1024,
.ucast_probes = 3,
.mcast_probes = 3,
.anycast_delay = 1 * HZ,
.proxy_delay = (8 * HZ) / 10,
.proxy_qlen = 64,
.data = {
[NEIGH_VAR_MCAST_PROBES] = 3,
[NEIGH_VAR_UCAST_PROBES] = 3,
[NEIGH_VAR_RETRANS_TIME] = ND_RETRANS_TIMER,
[NEIGH_VAR_BASE_REACHABLE_TIME] = ND_REACHABLE_TIME,
[NEIGH_VAR_DELAY_PROBE_TIME] = 5 * HZ,
[NEIGH_VAR_GC_STALETIME] = 60 * HZ,
[NEIGH_VAR_QUEUE_LEN_BYTES] = 64 * 1024,
[NEIGH_VAR_PROXY_QLEN] = 64,
[NEIGH_VAR_ANYCAST_DELAY] = 1 * HZ,
[NEIGH_VAR_PROXY_DELAY] = (8 * HZ) / 10,
},
},
.gc_interval = 30 * HZ,
.gc_thresh1 = 128,
......@@ -656,14 +658,14 @@ static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb)
if (skb && ipv6_chk_addr(dev_net(dev), &ipv6_hdr(skb)->saddr, dev, 1))
saddr = &ipv6_hdr(skb)->saddr;
if ((probes -= neigh->parms->ucast_probes) < 0) {
if ((probes -= NEIGH_VAR(neigh->parms, UCAST_PROBES)) < 0) {
if (!(neigh->nud_state & NUD_VALID)) {
ND_PRINTK(1, dbg,
"%s: trying to ucast probe in NUD_INVALID: %pI6\n",
__func__, target);
}
ndisc_send_ns(dev, neigh, target, target, saddr);
} else if ((probes -= neigh->parms->app_probes) < 0) {
} else if ((probes -= NEIGH_VAR(neigh->parms, APP_PROBES)) < 0) {
neigh_app_ns(neigh);
} else {
addrconf_addr_solict_mult(target, &mcaddr);
......@@ -790,7 +792,7 @@ static void ndisc_recv_ns(struct sk_buff *skb)
if (!(NEIGH_CB(skb)->flags & LOCALLY_ENQUEUED) &&
skb->pkt_type != PACKET_HOST &&
inc &&
idev->nd_parms->proxy_delay != 0) {
NEIGH_VAR(idev->nd_parms, PROXY_DELAY) != 0) {
/*
* for anycast or proxy,
* sender should delay its response
......@@ -1210,7 +1212,7 @@ static void ndisc_router_discovery(struct sk_buff *skb)
rtime = (rtime*HZ)/1000;
if (rtime < HZ/10)
rtime = HZ/10;
in6_dev->nd_parms->retrans_time = rtime;
NEIGH_VAR_SET(in6_dev->nd_parms, RETRANS_TIME, rtime);
in6_dev->tstamp = jiffies;
inet6_ifinfo_notify(RTM_NEWLINK, in6_dev);
}
......@@ -1222,9 +1224,11 @@ static void ndisc_router_discovery(struct sk_buff *skb)
if (rtime < HZ/10)
rtime = HZ/10;
if (rtime != in6_dev->nd_parms->base_reachable_time) {
in6_dev->nd_parms->base_reachable_time = rtime;
in6_dev->nd_parms->gc_staletime = 3 * rtime;
if (rtime != NEIGH_VAR(in6_dev->nd_parms, BASE_REACHABLE_TIME)) {
NEIGH_VAR_SET(in6_dev->nd_parms,
BASE_REACHABLE_TIME, rtime);
NEIGH_VAR_SET(in6_dev->nd_parms,
GC_STALETIME, 3 * rtime);
in6_dev->nd_parms->reachable_time = neigh_rand_reach_time(rtime);
in6_dev->tstamp = jiffies;
inet6_ifinfo_notify(RTM_NEWLINK, in6_dev);
......@@ -1651,22 +1655,23 @@ int ndisc_ifinfo_sysctl_change(struct ctl_table *ctl, int write, void __user *bu
ndisc_warn_deprecated_sysctl(ctl, "syscall", dev ? dev->name : "default");
if (strcmp(ctl->procname, "retrans_time") == 0)
ret = proc_dointvec(ctl, write, buffer, lenp, ppos);
ret = neigh_proc_dointvec(ctl, write, buffer, lenp, ppos);
else if (strcmp(ctl->procname, "base_reachable_time") == 0)
ret = proc_dointvec_jiffies(ctl, write,
buffer, lenp, ppos);
ret = neigh_proc_dointvec_jiffies(ctl, write,
buffer, lenp, ppos);
else if ((strcmp(ctl->procname, "retrans_time_ms") == 0) ||
(strcmp(ctl->procname, "base_reachable_time_ms") == 0))
ret = proc_dointvec_ms_jiffies(ctl, write,
buffer, lenp, ppos);
ret = neigh_proc_dointvec_ms_jiffies(ctl, write,
buffer, lenp, ppos);
else
ret = -1;
if (write && ret == 0 && dev && (idev = in6_dev_get(dev)) != NULL) {
if (ctl->data == &idev->nd_parms->base_reachable_time)
idev->nd_parms->reachable_time = neigh_rand_reach_time(idev->nd_parms->base_reachable_time);
if (ctl->data == &NEIGH_VAR(idev->nd_parms, BASE_REACHABLE_TIME))
idev->nd_parms->reachable_time =
neigh_rand_reach_time(NEIGH_VAR(idev->nd_parms, BASE_REACHABLE_TIME));
idev->tstamp = jiffies;
inet6_ifinfo_notify(RTM_NEWLINK, idev);
in6_dev_put(idev);
......@@ -1725,7 +1730,7 @@ int __init ndisc_init(void)
neigh_table_init(&nd_tbl);
#ifdef CONFIG_SYSCTL
err = neigh_sysctl_register(NULL, &nd_tbl.parms, "ipv6",
err = neigh_sysctl_register(NULL, &nd_tbl.parms,
&ndisc_ifinfo_sysctl_change);
if (err)
goto out_unregister_pernet;
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册