* [BK PATCH] [IPV6] Merge Specification Conformity Improvements
@ 2004-09-13 14:17 YOSHIFUJI Hideaki / 吉藤英明
2004-09-13 14:29 ` Pekka Savola
` (2 more replies)
0 siblings, 3 replies; 6+ messages in thread
From: YOSHIFUJI Hideaki / 吉藤英明 @ 2004-09-13 14:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yoshfuji, vnuorval
Hello.
Here're the changesets which improve conformity to
the IPv6 specifications / standards.
Please pull from
<bk://bk.skbuff.net:20609/linux-2.6-ndp-20040913/>
All changesets are product of USAGI/WIDE Project, and
signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>.
Plus, special thanks to Ville Nuorvala <vnuorval@tcs.hut.fi>
in GO/Core Project; credit is given in chagneset log.
We will make other couple of changesets for merging our efforts
by now.
Thank you!
HEADLINES
---------
ChangeSet@1.1868, 2004-09-13 15:50:00+09:00, yoshfuji@linux-ipv6.org
[NET] NEIGHBOUR: save number of arguments for neigh_update() by flags.
ChangeSet@1.1869, 2004-09-13 15:50:56+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: suspect REACHABLE entry if new lladdr is different.
ChangeSet@1.1870, 2004-09-13 15:51:20+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: keep original state if new state is STALE and lladdr is unchanged
ChangeSet@1.1871, 2004-09-13 15:52:13+09:00, yoshfuji@linux-ipv6.org
[NET] NEIGHBOUR: merge two flags for neigh_update() into one.
ChangeSet@1.1872, 2004-09-13 15:54:11+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: update IsRouter flag appropriately.
ChangeSet@1.1873, 2004-09-13 15:54:32+09:00, yoshfuji@linux-ipv6.org
[NET] NEIGHBOUR: use time_after() and its friends.
ChangeSet@1.1874, 2004-09-13 15:56:10+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: update entry appripriately when receiving NS.
ChangeSet@1.1875, 2004-09-13 15:56:55+09:00, yoshfuji@linux-ipv6.org
[NET] NEIGHBOUR: improve neighbour state machine.
ChangeSet@1.1876, 2004-09-13 15:57:40+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: Fix message validation against Redirects.
ChangeSet@1.1877, 2004-09-13 15:59:11+09:00, yoshfuji@linux-ipv6.org
[IPV6] don't use expired default routes.
ChangeSet@1.1878, 2004-09-13 16:02:20+09:00, yoshfuji@linux-ipv6.org
[IPV6] ensure to aging default routes.
ChangeSet@1.1879, 2004-09-13 16:04:16+09:00, yoshfuji@linux-ipv6.org
[IPV6] purge routes via non-router neighbour but gateway.
DIFFSTATS
---------
include/net/ip6_route.h | 1
include/net/neighbour.h | 5 -
net/core/neighbour.c | 221 ++++++++++++++++++++++++++----------------------
net/ipv6/ip6_fib.c | 12 ++
net/ipv6/ndisc.c | 56 ++++++------
net/ipv6/route.c | 54 +++++++----
6 files changed, 199 insertions(+), 150 deletions(-)
CHANGESETS
----------
ChangeSet@1.1868, 2004-09-13 15:50:00+09:00, yoshfuji@linux-ipv6.org
[NET] NEIGHBOUR: save number of arguments for neigh_update() by flags.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/include/net/neighbour.h b/include/net/neighbour.h
--- a/include/net/neighbour.h 2004-09-13 16:32:27 +09:00
+++ b/include/net/neighbour.h 2004-09-13 16:32:27 +09:00
@@ -179,6 +179,10 @@
struct pneigh_entry *phash_buckets[PNEIGH_HASHMASK+1];
};
+/* flags for neigh_update() */
+#define NEIGH_UPDATE_F_OVERRIDE 0x00000001
+#define NEIGH_UPDATE_F_ADMIN 0x80000000
+
extern void neigh_table_init(struct neigh_table *tbl);
extern int neigh_table_clear(struct neigh_table *tbl);
extern struct neighbour * neigh_lookup(struct neigh_table *tbl,
@@ -189,7 +193,8 @@
struct net_device *dev);
extern void neigh_destroy(struct neighbour *neigh);
extern int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb);
-extern int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new, int override, int arp);
+extern int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new,
+ u32 flags);
extern void neigh_changeaddr(struct neigh_table *tbl, struct net_device *dev);
extern int neigh_ifdown(struct neigh_table *tbl, struct net_device *dev);
extern int neigh_resolve_output(struct sk_buff *skb);
diff -Nru a/net/atm/clip.c b/net/atm/clip.c
--- a/net/atm/clip.c 2004-09-13 16:32:27 +09:00
+++ b/net/atm/clip.c 2004-09-13 16:32:27 +09:00
@@ -110,7 +110,8 @@
goto out;
entry->expires = jiffies-1;
/* force resolution or expiration */
- error = neigh_update(entry->neigh,NULL,NUD_NONE,0,0);
+ error = neigh_update(entry->neigh, NULL, NUD_NONE,
+ NEIGH_UPDATE_F_ADMIN);
if (error)
printk(KERN_CRIT "unlink_clip_vcc: "
"neigh_update failed with %d\n",error);
@@ -570,7 +571,8 @@
}
link_vcc(clip_vcc,entry);
}
- error = neigh_update(neigh,llc_oui,NUD_PERMANENT,1,0);
+ error = neigh_update(neigh, llc_oui, NUD_PERMANENT,
+ NEIGH_UPDATE_F_OVERRIDE|NEIGH_UPDATE_F_ADMIN);
neigh_release(neigh);
return error;
}
diff -Nru a/net/core/neighbour.c b/net/core/neighbour.c
--- a/net/core/neighbour.c 2004-09-13 16:32:27 +09:00
+++ b/net/core/neighbour.c 2004-09-13 16:32:27 +09:00
@@ -800,14 +800,16 @@
/* Generic update routine.
-- lladdr is new lladdr or NULL, if it is not supplied.
-- new is new state.
- -- override == 1 allows to override existing lladdr, if it is different.
- -- arp == 0 means that the change is administrative.
+ -- flags
+ NEIGH_UPDATE_F_OVERRIDE allows to override existing lladdr,
+ if it is different.
+ NEIGH_UPDATE_F_ADMIN means that the change is administrative.
Caller MUST hold reference count on the entry.
*/
int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new,
- int override, int arp)
+ u32 flags)
{
u8 old;
int err;
@@ -822,7 +824,8 @@
old = neigh->nud_state;
err = -EPERM;
- if (arp && (old & (NUD_NOARP | NUD_PERMANENT)))
+ if (!(flags & NEIGH_UPDATE_F_ADMIN) &&
+ (old & (NUD_NOARP | NUD_PERMANENT)))
goto out;
if (!(new & NUD_VALID)) {
@@ -850,7 +853,7 @@
if (old & NUD_VALID) {
if (!memcmp(lladdr, neigh->ha, dev->addr_len))
lladdr = neigh->ha;
- else if (!override)
+ else if (!(flags & NEIGH_UPDATE_F_OVERRIDE))
goto out;
}
} else {
@@ -928,7 +931,8 @@
struct neighbour *neigh = __neigh_lookup(tbl, saddr, dev,
lladdr || !dev->addr_len);
if (neigh)
- neigh_update(neigh, lladdr, NUD_STALE, 1, 1);
+ neigh_update(neigh, lladdr, NUD_STALE,
+ NEIGH_UPDATE_F_OVERRIDE);
return neigh;
}
@@ -1274,7 +1278,9 @@
n = neigh_lookup(tbl, RTA_DATA(nda[NDA_DST - 1]), dev);
if (n) {
- err = neigh_update(n, NULL, NUD_FAILED, 1, 0);
+ err = neigh_update(n, NULL, NUD_FAILED,
+ NEIGH_UPDATE_F_OVERRIDE|
+ NEIGH_UPDATE_F_ADMIN);
neigh_release(n);
}
goto out_dev_put;
@@ -1347,7 +1353,8 @@
RTA_DATA(nda[NDA_LLADDR - 1]) :
NULL,
ndm->ndm_state,
- override, 0);
+ (override ? NEIGH_UPDATE_F_OVERRIDE : 0) |
+ NEIGH_UPDATE_F_ADMIN);
}
if (n)
neigh_release(n);
diff -Nru a/net/ipv4/arp.c b/net/ipv4/arp.c
--- a/net/ipv4/arp.c 2004-09-13 16:32:27 +09:00
+++ b/net/ipv4/arp.c 2004-09-13 16:32:27 +09:00
@@ -914,7 +914,7 @@
if (arp->ar_op != htons(ARPOP_REPLY) ||
skb->pkt_type != PACKET_HOST)
state = NUD_STALE;
- neigh_update(n, sha, state, override, 1);
+ neigh_update(n, sha, state, override ? NEIGH_UPDATE_F_OVERRIDE : 0);
neigh_release(n);
}
@@ -1021,7 +1021,9 @@
if (r->arp_flags & ATF_PERM)
state = NUD_PERMANENT;
err = neigh_update(neigh, (r->arp_flags&ATF_COM) ?
- r->arp_ha.sa_data : NULL, state, 1, 0);
+ r->arp_ha.sa_data : NULL, state,
+ NEIGH_UPDATE_F_OVERRIDE|
+ NEIGH_UPDATE_F_ADMIN);
neigh_release(neigh);
}
return err;
@@ -1101,7 +1103,9 @@
neigh = neigh_lookup(&arp_tbl, &ip, dev);
if (neigh) {
if (neigh->nud_state&~NUD_NOARP)
- err = neigh_update(neigh, NULL, NUD_FAILED, 1, 0);
+ err = neigh_update(neigh, NULL, NUD_FAILED,
+ NEIGH_UPDATE_F_OVERRIDE|
+ NEIGH_UPDATE_F_ADMIN);
neigh_release(neigh);
}
return err;
diff -Nru a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
--- a/net/ipv6/ndisc.c 2004-09-13 16:32:27 +09:00
+++ b/net/ipv6/ndisc.c 2004-09-13 16:32:27 +09:00
@@ -911,7 +911,7 @@
neigh_update(neigh, lladdr,
msg->icmph.icmp6_solicited ? NUD_REACHABLE : NUD_STALE,
- msg->icmph.icmp6_override, 1);
+ msg->icmph.icmp6_override ? NEIGH_UPDATE_F_OVERRIDE : 0);
neigh_release(neigh);
}
}
@@ -1079,7 +1079,7 @@
goto out;
}
}
- neigh_update(neigh, lladdr, NUD_STALE, 1, 1);
+ neigh_update(neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_OVERRIDE);
}
if (ndopts.nd_opts_pi) {
@@ -1204,7 +1204,7 @@
neigh = __neigh_lookup(&nd_tbl, target, skb->dev, 1);
if (neigh) {
- neigh_update(neigh, lladdr, NUD_STALE, 1, 1);
+ neigh_update(neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_OVERRIDE);
if (neigh->nud_state&NUD_VALID)
rt6_redirect(dest, &skb->nh.ipv6h->saddr, neigh, on_link);
else
ChangeSet@1.1869, 2004-09-13 15:50:56+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: suspect REACHABLE entry if new lladdr is different.
When we receive NA without Override flag, if it comes with
different lladdr from one in our REACHABLE entry,
set the state to STALE. (RFC2461 7.2.5)
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/include/net/neighbour.h b/include/net/neighbour.h
--- a/include/net/neighbour.h 2004-09-13 16:32:31 +09:00
+++ b/include/net/neighbour.h 2004-09-13 16:32:31 +09:00
@@ -181,6 +181,7 @@
/* flags for neigh_update() */
#define NEIGH_UPDATE_F_OVERRIDE 0x00000001
+#define NEIGH_UPDATE_F_SUSPECT_CONNECTED 0x00000002
#define NEIGH_UPDATE_F_ADMIN 0x80000000
extern void neigh_table_init(struct neigh_table *tbl);
diff -Nru a/net/core/neighbour.c b/net/core/neighbour.c
--- a/net/core/neighbour.c 2004-09-13 16:32:31 +09:00
+++ b/net/core/neighbour.c 2004-09-13 16:32:31 +09:00
@@ -803,6 +803,9 @@
-- flags
NEIGH_UPDATE_F_OVERRIDE allows to override existing lladdr,
if it is different.
+ NEIGH_UPDATE_F_SUSPECT_CONNECTED will suspect existing "connected"
+ lladdr instead of overriding it
+ if it is different.
NEIGH_UPDATE_F_ADMIN means that the change is administrative.
Caller MUST hold reference count on the entry.
@@ -850,12 +853,9 @@
- compare new & old
- if they are different, check override flag
*/
- if (old & NUD_VALID) {
- if (!memcmp(lladdr, neigh->ha, dev->addr_len))
- lladdr = neigh->ha;
- else if (!(flags & NEIGH_UPDATE_F_OVERRIDE))
- goto out;
- }
+ if ((old & NUD_VALID) &&
+ !memcmp(lladdr, neigh->ha, dev->addr_len))
+ lladdr = neigh->ha;
} else {
/* No address is supplied; if we know something,
use it, otherwise discard the request.
@@ -876,9 +876,20 @@
do not change entry state, if new one is STALE.
*/
err = 0;
- if ((old & NUD_VALID) && lladdr == neigh->ha &&
- (new == old || (new == NUD_STALE && (old & NUD_CONNECTED))))
- goto out;
+ if (old & NUD_VALID) {
+ if (lladdr != neigh->ha && !(flags & NEIGH_UPDATE_F_OVERRIDE)) {
+ if ((flags & NEIGH_UPDATE_F_SUSPECT_CONNECTED) &&
+ (old & NUD_CONNECTED)) {
+ lladdr = neigh->ha;
+ new = NUD_STALE;
+ } else
+ goto out;
+ } else {
+ if (lladdr == neigh->ha &&
+ new == NUD_STALE && (old & NUD_CONNECTED))
+ new = old;
+ }
+ }
neigh_del_timer(neigh);
neigh->nud_state = new;
diff -Nru a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
--- a/net/ipv6/ndisc.c 2004-09-13 16:32:31 +09:00
+++ b/net/ipv6/ndisc.c 2004-09-13 16:32:31 +09:00
@@ -911,7 +911,8 @@
neigh_update(neigh, lladdr,
msg->icmph.icmp6_solicited ? NUD_REACHABLE : NUD_STALE,
- msg->icmph.icmp6_override ? NEIGH_UPDATE_F_OVERRIDE : 0);
+ NEIGH_UPDATE_F_SUSPECT_CONNECTED|
+ (msg->icmph.icmp6_override ? NEIGH_UPDATE_F_OVERRIDE : 0));
neigh_release(neigh);
}
}
ChangeSet@1.1870, 2004-09-13 15:51:20+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: keep original state if new state is STALE and lladdr is unchanged
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/include/net/neighbour.h b/include/net/neighbour.h
--- a/include/net/neighbour.h 2004-09-13 16:32:34 +09:00
+++ b/include/net/neighbour.h 2004-09-13 16:32:34 +09:00
@@ -182,6 +182,7 @@
/* flags for neigh_update() */
#define NEIGH_UPDATE_F_OVERRIDE 0x00000001
#define NEIGH_UPDATE_F_SUSPECT_CONNECTED 0x00000002
+#define NEIGH_UPDATE_F_RETAIN_STATE 0x00000004
#define NEIGH_UPDATE_F_ADMIN 0x80000000
extern void neigh_table_init(struct neigh_table *tbl);
diff -Nru a/net/core/neighbour.c b/net/core/neighbour.c
--- a/net/core/neighbour.c 2004-09-13 16:32:34 +09:00
+++ b/net/core/neighbour.c 2004-09-13 16:32:34 +09:00
@@ -806,6 +806,8 @@
NEIGH_UPDATE_F_SUSPECT_CONNECTED will suspect existing "connected"
lladdr instead of overriding it
if it is different.
+ NEIGH_UPDATE_F_RETAIN_STATE allows to retain current state
+ if lladdr is unchanged.
NEIGH_UPDATE_F_ADMIN means that the change is administrative.
Caller MUST hold reference count on the entry.
@@ -885,8 +887,9 @@
} else
goto out;
} else {
- if (lladdr == neigh->ha &&
- new == NUD_STALE && (old & NUD_CONNECTED))
+ if (lladdr == neigh->ha && new == NUD_STALE &&
+ ((flags & NEIGH_UPDATE_F_RETAIN_STATE) ||
+ (old & NUD_CONNECTED)))
new = old;
}
}
diff -Nru a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
--- a/net/ipv6/ndisc.c 2004-09-13 16:32:34 +09:00
+++ b/net/ipv6/ndisc.c 2004-09-13 16:32:34 +09:00
@@ -912,6 +912,7 @@
neigh_update(neigh, lladdr,
msg->icmph.icmp6_solicited ? NUD_REACHABLE : NUD_STALE,
NEIGH_UPDATE_F_SUSPECT_CONNECTED|
+ NEIGH_UPDATE_F_RETAIN_STATE|
(msg->icmph.icmp6_override ? NEIGH_UPDATE_F_OVERRIDE : 0));
neigh_release(neigh);
}
@@ -1080,7 +1081,9 @@
goto out;
}
}
- neigh_update(neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_OVERRIDE);
+ neigh_update(neigh, lladdr, NUD_STALE,
+ NEIGH_UPDATE_F_RETAIN_STATE|
+ NEIGH_UPDATE_F_OVERRIDE);
}
if (ndopts.nd_opts_pi) {
@@ -1205,7 +1208,9 @@
neigh = __neigh_lookup(&nd_tbl, target, skb->dev, 1);
if (neigh) {
- neigh_update(neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_OVERRIDE);
+ neigh_update(neigh, lladdr, NUD_STALE,
+ NEIGH_UPDATE_F_RETAIN_STATE|
+ NEIGH_UPDATE_F_OVERRIDE);
if (neigh->nud_state&NUD_VALID)
rt6_redirect(dest, &skb->nh.ipv6h->saddr, neigh, on_link);
else
ChangeSet@1.1871, 2004-09-13 15:52:13+09:00, yoshfuji@linux-ipv6.org
[NET] NEIGHBOUR: merge two flags for neigh_update() into one.
This is because SUSPECT_CONNECTED can be effective
only if OVERRIDE is unset, and used only if RETAIN_STATE is set.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/include/net/neighbour.h b/include/net/neighbour.h
--- a/include/net/neighbour.h 2004-09-13 16:32:37 +09:00
+++ b/include/net/neighbour.h 2004-09-13 16:32:37 +09:00
@@ -181,8 +181,7 @@
/* flags for neigh_update() */
#define NEIGH_UPDATE_F_OVERRIDE 0x00000001
-#define NEIGH_UPDATE_F_SUSPECT_CONNECTED 0x00000002
-#define NEIGH_UPDATE_F_RETAIN_STATE 0x00000004
+#define NEIGH_UPDATE_F_WEAK_OVERRIDE 0x00000002
#define NEIGH_UPDATE_F_ADMIN 0x80000000
extern void neigh_table_init(struct neigh_table *tbl);
diff -Nru a/net/core/neighbour.c b/net/core/neighbour.c
--- a/net/core/neighbour.c 2004-09-13 16:32:37 +09:00
+++ b/net/core/neighbour.c 2004-09-13 16:32:37 +09:00
@@ -803,10 +803,10 @@
-- flags
NEIGH_UPDATE_F_OVERRIDE allows to override existing lladdr,
if it is different.
- NEIGH_UPDATE_F_SUSPECT_CONNECTED will suspect existing "connected"
+ NEIGH_UPDATE_F_WEAK_OVERRIDE will suspect existing "connected"
lladdr instead of overriding it
if it is different.
- NEIGH_UPDATE_F_RETAIN_STATE allows to retain current state
+ It also allows to retain current state
if lladdr is unchanged.
NEIGH_UPDATE_F_ADMIN means that the change is administrative.
@@ -880,7 +880,7 @@
err = 0;
if (old & NUD_VALID) {
if (lladdr != neigh->ha && !(flags & NEIGH_UPDATE_F_OVERRIDE)) {
- if ((flags & NEIGH_UPDATE_F_SUSPECT_CONNECTED) &&
+ if ((flags & NEIGH_UPDATE_F_WEAK_OVERRIDE) &&
(old & NUD_CONNECTED)) {
lladdr = neigh->ha;
new = NUD_STALE;
@@ -888,8 +888,9 @@
goto out;
} else {
if (lladdr == neigh->ha && new == NUD_STALE &&
- ((flags & NEIGH_UPDATE_F_RETAIN_STATE) ||
- (old & NUD_CONNECTED)))
+ ((flags & NEIGH_UPDATE_F_WEAK_OVERRIDE) ||
+ (old & NUD_CONNECTED))
+ )
new = old;
}
}
diff -Nru a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
--- a/net/ipv6/ndisc.c 2004-09-13 16:32:37 +09:00
+++ b/net/ipv6/ndisc.c 2004-09-13 16:32:37 +09:00
@@ -911,8 +911,7 @@
neigh_update(neigh, lladdr,
msg->icmph.icmp6_solicited ? NUD_REACHABLE : NUD_STALE,
- NEIGH_UPDATE_F_SUSPECT_CONNECTED|
- NEIGH_UPDATE_F_RETAIN_STATE|
+ NEIGH_UPDATE_F_WEAK_OVERRIDE|
(msg->icmph.icmp6_override ? NEIGH_UPDATE_F_OVERRIDE : 0));
neigh_release(neigh);
}
@@ -1082,7 +1081,7 @@
}
}
neigh_update(neigh, lladdr, NUD_STALE,
- NEIGH_UPDATE_F_RETAIN_STATE|
+ NEIGH_UPDATE_F_WEAK_OVERRIDE|
NEIGH_UPDATE_F_OVERRIDE);
}
@@ -1209,7 +1208,7 @@
neigh = __neigh_lookup(&nd_tbl, target, skb->dev, 1);
if (neigh) {
neigh_update(neigh, lladdr, NUD_STALE,
- NEIGH_UPDATE_F_RETAIN_STATE|
+ NEIGH_UPDATE_F_WEAK_OVERRIDE|
NEIGH_UPDATE_F_OVERRIDE);
if (neigh->nud_state&NUD_VALID)
rt6_redirect(dest, &skb->nh.ipv6h->saddr, neigh, on_link);
ChangeSet@1.1872, 2004-09-13 15:54:11+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: update IsRouter flag appropriately.
Update IsRouter (NTF_ROUTER) flag approrpriately.
Specifically,
- we should not update it blindly; if Override Flag is
unset and lladdr is differnt, we should NOT.
- we should set it when we have received RA.
- we should set it when we have received Redirect
whose target is off-link.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/include/net/neighbour.h b/include/net/neighbour.h
--- a/include/net/neighbour.h 2004-09-13 16:32:40 +09:00
+++ b/include/net/neighbour.h 2004-09-13 16:32:40 +09:00
@@ -182,6 +182,8 @@
/* flags for neigh_update() */
#define NEIGH_UPDATE_F_OVERRIDE 0x00000001
#define NEIGH_UPDATE_F_WEAK_OVERRIDE 0x00000002
+#define NEIGH_UPDATE_F_OVERRIDE_ISROUTER 0x00000004
+#define NEIGH_UPDATE_F_ISROUTER 0x40000000
#define NEIGH_UPDATE_F_ADMIN 0x80000000
extern void neigh_table_init(struct neigh_table *tbl);
diff -Nru a/net/core/neighbour.c b/net/core/neighbour.c
--- a/net/core/neighbour.c 2004-09-13 16:32:40 +09:00
+++ b/net/core/neighbour.c 2004-09-13 16:32:40 +09:00
@@ -810,6 +810,11 @@
if lladdr is unchanged.
NEIGH_UPDATE_F_ADMIN means that the change is administrative.
+ NEIGH_UPDATE_F_OVERRIDE_ISROUTER allows to override existing
+ NTF_ROUTER flag.
+ NEIGH_UPDATE_F_ISROUTER indicates if the neighbour is known as
+ a router.
+
Caller MUST hold reference count on the entry.
*/
@@ -822,6 +827,7 @@
int notify = 0;
#endif
struct net_device *dev;
+ int update_isrouter = 0;
write_lock_bh(&neigh->lock);
@@ -878,8 +884,10 @@
do not change entry state, if new one is STALE.
*/
err = 0;
+ update_isrouter = flags & NEIGH_UPDATE_F_OVERRIDE_ISROUTER;
if (old & NUD_VALID) {
if (lladdr != neigh->ha && !(flags & NEIGH_UPDATE_F_OVERRIDE)) {
+ update_isrouter = 0;
if ((flags & NEIGH_UPDATE_F_WEAK_OVERRIDE) &&
(old & NUD_CONNECTED)) {
lladdr = neigh->ha;
@@ -931,6 +939,11 @@
skb_queue_purge(&neigh->arp_queue);
}
out:
+ if (update_isrouter) {
+ neigh->flags = (flags & NEIGH_UPDATE_F_ISROUTER) ?
+ (neigh->flags | NTF_ROUTER) :
+ (neigh->flags & ~NTF_ROUTER);
+ }
write_unlock_bh(&neigh->lock);
#ifdef CONFIG_ARPD
if (notify && neigh->parms->app_probes)
diff -Nru a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
--- a/net/ipv6/ndisc.c 2004-09-13 16:32:40 +09:00
+++ b/net/ipv6/ndisc.c 2004-09-13 16:32:40 +09:00
@@ -894,25 +894,25 @@
neigh = neigh_lookup(&nd_tbl, &msg->target, dev);
if (neigh) {
- if (neigh->flags & NTF_ROUTER) {
- if (msg->icmph.icmp6_router == 0) {
- /*
- * Change: router to host
- */
- struct rt6_info *rt;
- rt = rt6_get_dflt_router(saddr, dev);
- if (rt)
- ip6_del_rt(rt, NULL, NULL);
- }
- } else {
- if (msg->icmph.icmp6_router)
- neigh->flags |= NTF_ROUTER;
- }
+ u8 old_flags = neigh->flags;
neigh_update(neigh, lladdr,
msg->icmph.icmp6_solicited ? NUD_REACHABLE : NUD_STALE,
NEIGH_UPDATE_F_WEAK_OVERRIDE|
- (msg->icmph.icmp6_override ? NEIGH_UPDATE_F_OVERRIDE : 0));
+ (msg->icmph.icmp6_override ? NEIGH_UPDATE_F_OVERRIDE : 0)|
+ NEIGH_UPDATE_F_OVERRIDE_ISROUTER|
+ (msg->icmph.icmp6_router ? NEIGH_UPDATE_F_ISROUTER : 0));
+
+ if ((old_flags & ~neigh->flags) & NTF_ROUTER) {
+ /*
+ * Change: router to host
+ */
+ struct rt6_info *rt;
+ rt = rt6_get_dflt_router(saddr, dev);
+ if (rt)
+ ip6_del_rt(rt, NULL, NULL);
+ }
+
neigh_release(neigh);
}
}
@@ -1082,7 +1082,9 @@
}
neigh_update(neigh, lladdr, NUD_STALE,
NEIGH_UPDATE_F_WEAK_OVERRIDE|
- NEIGH_UPDATE_F_OVERRIDE);
+ NEIGH_UPDATE_F_OVERRIDE|
+ NEIGH_UPDATE_F_OVERRIDE_ISROUTER|
+ NEIGH_UPDATE_F_ISROUTER);
}
if (ndopts.nd_opts_pi) {
@@ -1209,7 +1211,10 @@
if (neigh) {
neigh_update(neigh, lladdr, NUD_STALE,
NEIGH_UPDATE_F_WEAK_OVERRIDE|
- NEIGH_UPDATE_F_OVERRIDE);
+ NEIGH_UPDATE_F_OVERRIDE|
+ (on_link ? 0 : (NEIGH_UPDATE_F_OVERRIDE_ISROUTER|
+ NEIGH_UPDATE_F_ISROUTER))
+ );
if (neigh->nud_state&NUD_VALID)
rt6_redirect(dest, &skb->nh.ipv6h->saddr, neigh, on_link);
else
ChangeSet@1.1873, 2004-09-13 15:54:32+09:00, yoshfuji@linux-ipv6.org
[NET] NEIGHBOUR: use time_after() and its friends.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/net/core/neighbour.c b/net/core/neighbour.c
--- a/net/core/neighbour.c 2004-09-13 16:32:43 +09:00
+++ b/net/core/neighbour.c 2004-09-13 16:32:43 +09:00
@@ -133,7 +133,7 @@
if (atomic_read(&n->refcnt) == 1 &&
!(n->nud_state & NUD_PERMANENT) &&
(n->nud_state != NUD_INCOMPLETE ||
- jiffies - n->used > n->parms->retrans_time)) {
+ time_after(jiffies, n->used + n->parms->retrans_time))) {
*np = n->next;
n->dead = 1;
shrunk = 1;
@@ -255,7 +255,7 @@
if (tbl->entries > tbl->gc_thresh3 ||
(tbl->entries > tbl->gc_thresh2 &&
- now - tbl->last_flush > 5 * HZ)) {
+ time_after(now, tbl->last_flush + 5 * HZ))) {
if (!neigh_forced_gc(tbl) &&
tbl->entries > tbl->gc_thresh3)
goto out;
@@ -563,12 +563,12 @@
if (state & (NUD_NOARP | NUD_PERMANENT))
return;
if (state & NUD_REACHABLE) {
- if (now - n->confirmed > n->parms->reachable_time) {
+ if (time_after(now, n->confirmed + n->parms->reachable_time)) {
n->nud_state = NUD_STALE;
neigh_suspect(n);
}
} else if (state & NUD_VALID) {
- if (now - n->confirmed < n->parms->reachable_time) {
+ if (time_before(now, n->confirmed + n->parms->reachable_time)) {
neigh_del_timer(n);
n->nud_state = NUD_REACHABLE;
neigh_connect(n);
@@ -589,7 +589,7 @@
* periodically recompute ReachableTime from random function
*/
- if (now - tbl->last_rand > 300 * HZ) {
+ if (time_after(now, tbl->last_rand + 300 * HZ)) {
struct neigh_parms *p;
tbl->last_rand = now;
for (p = &tbl->parms; p; p = p->next)
@@ -612,12 +612,12 @@
goto next_elt;
}
- if ((long)(n->used - n->confirmed) < 0)
+ if (time_before(n->used, n->confirmed))
n->used = n->confirmed;
if (atomic_read(&n->refcnt) == 1 &&
(state == NUD_FAILED ||
- now - n->used > n->parms->gc_staletime)) {
+ time_after(now, n->used + n->parms->gc_staletime))) {
*np = n->next;
n->dead = 1;
write_unlock(&n->lock);
@@ -626,7 +626,7 @@
}
if (n->nud_state & NUD_REACHABLE &&
- now - n->confirmed > n->parms->reachable_time) {
+ time_after(now, n->confirmed + n->parms->reachable_time)) {
n->nud_state = NUD_STALE;
neigh_suspect(n);
}
@@ -671,7 +671,7 @@
}
if ((state & NUD_VALID) &&
- now - neigh->confirmed < neigh->parms->reachable_time) {
+ time_before(now, neigh->confirmed + neigh->parms->reachable_time)) {
neigh->nud_state = NUD_REACHABLE;
NEIGH_PRINTK2("neigh %p is still alive.\n", neigh);
neigh_connect(neigh);
@@ -1126,26 +1126,25 @@
struct sk_buff *skb)
{
unsigned long now = jiffies;
- long sched_next = net_random() % p->proxy_delay;
+ unsigned long sched_next = now + (net_random() % p->proxy_delay);
if (tbl->proxy_queue.qlen > p->proxy_qlen) {
kfree_skb(skb);
return;
}
skb->stamp.tv_sec = LOCALLY_ENQUEUED;
- skb->stamp.tv_usec = now + sched_next;
+ skb->stamp.tv_usec = sched_next;
spin_lock(&tbl->proxy_queue.lock);
if (del_timer(&tbl->proxy_timer)) {
- long tval = tbl->proxy_timer.expires - now;
- if (tval < sched_next)
- sched_next = tval;
+ if (time_before(tbl->proxy_timer.expires, sched_next))
+ sched_next = tbl->proxy_timer.expires;
}
dst_release(skb->dst);
skb->dst = NULL;
dev_hold(skb->dev);
__skb_queue_tail(&tbl->proxy_queue, skb);
- mod_timer(&tbl->proxy_timer, now + sched_next);
+ mod_timer(&tbl->proxy_timer, sched_next);
spin_unlock(&tbl->proxy_queue.lock);
}
ChangeSet@1.1874, 2004-09-13 15:56:10+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: update entry appripriately when receiving NS.
Update neighbour entry appropriately by passing correct flags
when receiving NS.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
--- a/net/ipv6/ndisc.c 2004-09-13 16:32:46 +09:00
+++ b/net/ipv6/ndisc.c 2004-09-13 16:32:46 +09:00
@@ -810,8 +810,11 @@
* update / create cache entry
* for the source address
*/
- neigh = neigh_event_ns(&nd_tbl, lladdr, saddr, dev);
-
+ neigh = __neigh_lookup(&nd_tbl, saddr, dev, lladdr || !dev->addr_len);
+ if (neigh)
+ neigh_update(neigh, lladdr, NUD_STALE,
+ NEIGH_UPDATE_F_WEAK_OVERRIDE|
+ NEIGH_UPDATE_F_OVERRIDE);
if (neigh || !dev->hard_header) {
ndisc_send_na(dev, neigh, saddr, &msg->target,
idev->cnf.forwarding,
ChangeSet@1.1875, 2004-09-13 15:56:55+09:00, yoshfuji@linux-ipv6.org
[NET] NEIGHBOUR: improve neighbour state machine.
This centralizes neighbour state transition by timer into
neigh_timer_handler(), and kill neigh_sync().
This improves timing accuracy of state transition.
neigh_timer_handler() for each entry is now reponsible
for state transition of the entry, and
neigh_periodic_timer() is just for garbage collection.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/include/net/neighbour.h b/include/net/neighbour.h
--- a/include/net/neighbour.h 2004-09-13 16:32:50 +09:00
+++ b/include/net/neighbour.h 2004-09-13 16:32:50 +09:00
@@ -51,7 +51,7 @@
#include <linux/err.h>
#include <linux/sysctl.h>
-#define NUD_IN_TIMER (NUD_INCOMPLETE|NUD_DELAY|NUD_PROBE)
+#define NUD_IN_TIMER (NUD_INCOMPLETE|NUD_REACHABLE|NUD_DELAY|NUD_PROBE)
#define NUD_VALID (NUD_PERMANENT|NUD_NOARP|NUD_REACHABLE|NUD_PROBE|NUD_STALE|NUD_DELAY)
#define NUD_CONNECTED (NUD_PERMANENT|NUD_NOARP|NUD_REACHABLE)
diff -Nru a/net/core/neighbour.c b/net/core/neighbour.c
--- a/net/core/neighbour.c 2004-09-13 16:32:50 +09:00
+++ b/net/core/neighbour.c 2004-09-13 16:32:50 +09:00
@@ -542,40 +542,6 @@
hh->hh_output = neigh->ops->hh_output;
}
-/*
- Transitions NUD_STALE <-> NUD_REACHABLE do not occur
- when fast path is built: we have no timers associated with
- these states, we do not have time to check state when sending.
- neigh_periodic_timer check periodically neigh->confirmed
- time and moves NUD_REACHABLE -> NUD_STALE.
-
- If a routine wants to know TRUE entry state, it calls
- neigh_sync before checking state.
-
- Called with write_locked neigh.
- */
-
-static void neigh_sync(struct neighbour *n)
-{
- unsigned long now = jiffies;
- u8 state = n->nud_state;
-
- if (state & (NUD_NOARP | NUD_PERMANENT))
- return;
- if (state & NUD_REACHABLE) {
- if (time_after(now, n->confirmed + n->parms->reachable_time)) {
- n->nud_state = NUD_STALE;
- neigh_suspect(n);
- }
- } else if (state & NUD_VALID) {
- if (time_before(now, n->confirmed + n->parms->reachable_time)) {
- neigh_del_timer(n);
- n->nud_state = NUD_REACHABLE;
- neigh_connect(n);
- }
- }
-}
-
static void neigh_periodic_timer(unsigned long arg)
{
struct neigh_table *tbl = (struct neigh_table *)arg;
@@ -624,12 +590,6 @@
neigh_release(n);
continue;
}
-
- if (n->nud_state & NUD_REACHABLE &&
- time_after(now, n->confirmed + n->parms->reachable_time)) {
- n->nud_state = NUD_STALE;
- neigh_suspect(n);
- }
write_unlock(&n->lock);
next_elt:
@@ -654,7 +614,7 @@
static void neigh_timer_handler(unsigned long arg)
{
- unsigned long now = jiffies;
+ unsigned long now, next;
struct neighbour *neigh = (struct neighbour *)arg;
unsigned state;
int notify = 0;
@@ -662,6 +622,8 @@
write_lock(&neigh->lock);
state = neigh->nud_state;
+ now = jiffies;
+ next = now + HZ;
if (!(state & NUD_IN_TIMER)) {
#ifndef CONFIG_SMP
@@ -670,20 +632,42 @@
goto out;
}
- if ((state & NUD_VALID) &&
- time_before(now, neigh->confirmed + neigh->parms->reachable_time)) {
- neigh->nud_state = NUD_REACHABLE;
- NEIGH_PRINTK2("neigh %p is still alive.\n", neigh);
- neigh_connect(neigh);
- goto out;
- }
- if (state == NUD_DELAY) {
- NEIGH_PRINTK2("neigh %p is probed.\n", neigh);
- neigh->nud_state = NUD_PROBE;
- atomic_set(&neigh->probes, 0);
+ if (state & NUD_REACHABLE) {
+ if (time_before_eq(now,
+ neigh->confirmed + neigh->parms->reachable_time)) {
+ NEIGH_PRINTK2("neigh %p is still alive.\n", neigh);
+ next = neigh->confirmed + neigh->parms->reachable_time;
+ } else if (time_before_eq(now,
+ neigh->used + neigh->parms->delay_probe_time)) {
+ NEIGH_PRINTK2("neigh %p is delayed.\n", neigh);
+ neigh->nud_state = NUD_DELAY;
+ neigh_suspect(neigh);
+ next = now + neigh->parms->delay_probe_time;
+ } else {
+ NEIGH_PRINTK2("neigh %p is suspected.\n", neigh);
+ neigh->nud_state = NUD_STALE;
+ neigh_suspect(neigh);
+ }
+ } else if (state & NUD_DELAY) {
+ if (time_before_eq(now,
+ neigh->confirmed + neigh->parms->delay_probe_time)) {
+ NEIGH_PRINTK2("neigh %p is now reachable.\n", neigh);
+ neigh->nud_state = NUD_REACHABLE;
+ neigh_connect(neigh);
+ next = neigh->confirmed + neigh->parms->reachable_time;
+ } else {
+ NEIGH_PRINTK2("neigh %p is probed.\n", neigh);
+ neigh->nud_state = NUD_PROBE;
+ atomic_set(&neigh->probes, 0);
+ next = now + neigh->parms->retrans_time;
+ }
+ } else {
+ /* NUD_PROBE|NUD_INCOMPLETE */
+ next = now + neigh->parms->retrans_time;
}
- if (atomic_read(&neigh->probes) >= neigh_max_probes(neigh)) {
+ if ((neigh->nud_state & (NUD_INCOMPLETE | NUD_PROBE)) &&
+ atomic_read(&neigh->probes) >= neigh_max_probes(neigh)) {
struct sk_buff *skb;
neigh->nud_state = NUD_FAILED;
@@ -703,19 +687,24 @@
write_lock(&neigh->lock);
}
skb_queue_purge(&neigh->arp_queue);
- goto out;
}
- neigh->timer.expires = now + neigh->parms->retrans_time;
- add_timer(&neigh->timer);
- write_unlock(&neigh->lock);
-
- neigh->ops->solicit(neigh, skb_peek(&neigh->arp_queue));
- atomic_inc(&neigh->probes);
- return;
-
+ if (neigh->nud_state & NUD_IN_TIMER) {
+ neigh_hold(neigh);
+ if (time_before(next, jiffies + HZ/2))
+ next = jiffies + HZ/2;
+ neigh->timer.expires = next;
+ add_timer(&neigh->timer);
+ }
+ if (neigh->nud_state & (NUD_INCOMPLETE | NUD_PROBE)) {
+ write_unlock(&neigh->lock);
+ neigh->ops->solicit(neigh, skb_peek(&neigh->arp_queue));
+ atomic_inc(&neigh->probes);
+ } else {
out:
- write_unlock(&neigh->lock);
+ write_unlock(&neigh->lock);
+ }
+
#ifdef CONFIG_ARPD
if (notify && neigh->parms->app_probes)
neigh_app_notify(neigh);
@@ -726,6 +715,7 @@
int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb)
{
int rc;
+ unsigned long now;
write_lock_bh(&neigh->lock);
@@ -733,18 +723,15 @@
if (neigh->nud_state & (NUD_CONNECTED | NUD_DELAY | NUD_PROBE))
goto out_unlock_bh;
+ now = jiffies;
+
if (!(neigh->nud_state & (NUD_STALE | NUD_INCOMPLETE))) {
if (neigh->parms->mcast_probes + neigh->parms->app_probes) {
atomic_set(&neigh->probes, neigh->parms->ucast_probes);
neigh->nud_state = NUD_INCOMPLETE;
neigh_hold(neigh);
- neigh->timer.expires = jiffies +
- neigh->parms->retrans_time;
+ neigh->timer.expires = now + 1;
add_timer(&neigh->timer);
- write_unlock_bh(&neigh->lock);
- neigh->ops->solicit(neigh, skb);
- atomic_inc(&neigh->probes);
- write_lock_bh(&neigh->lock);
} else {
neigh->nud_state = NUD_FAILED;
write_unlock_bh(&neigh->lock);
@@ -753,6 +740,12 @@
kfree_skb(skb);
return 1;
}
+ } else if (neigh->nud_state & NUD_STALE) {
+ NEIGH_PRINTK2("neigh %p is delayed.\n", neigh);
+ neigh_hold(neigh);
+ neigh->nud_state = NUD_DELAY;
+ neigh->timer.expires = jiffies + neigh->parms->delay_probe_time;
+ add_timer(&neigh->timer);
}
if (neigh->nud_state == NUD_INCOMPLETE) {
@@ -767,13 +760,6 @@
__skb_queue_tail(&neigh->arp_queue, skb);
}
rc = 1;
- } else if (neigh->nud_state == NUD_STALE) {
- NEIGH_PRINTK2("neigh %p is delayed.\n", neigh);
- neigh_hold(neigh);
- neigh->nud_state = NUD_DELAY;
- neigh->timer.expires = jiffies + neigh->parms->delay_probe_time;
- add_timer(&neigh->timer);
- rc = 0;
}
out_unlock_bh:
write_unlock_bh(&neigh->lock);
@@ -874,8 +860,6 @@
lladdr = neigh->ha;
}
- neigh_sync(neigh);
- old = neigh->nud_state;
if (new & NUD_CONNECTED)
neigh->confirmed = jiffies;
neigh->updated = jiffies;
@@ -903,8 +887,18 @@
}
}
- neigh_del_timer(neigh);
- neigh->nud_state = new;
+ if (new != old) {
+ neigh_del_timer(neigh);
+ if (new & NUD_IN_TIMER) {
+ neigh_hold(neigh);
+ neigh->timer.expires = jiffies +
+ ((new & NUD_REACHABLE) ?
+ neigh->parms->reachable_time : 0);
+ add_timer(&neigh->timer);
+ }
+ neigh->nud_state = new;
+ }
+
if (lladdr != neigh->ha) {
memcpy(&neigh->ha, lladdr, dev->addr_len);
neigh_update_hhs(neigh);
ChangeSet@1.1876, 2004-09-13 15:57:40+09:00, yoshfuji@linux-ipv6.org
[IPV6] NDISC: Fix message validation against Redirects.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/include/net/ip6_route.h b/include/net/ip6_route.h
--- a/include/net/ip6_route.h 2004-09-13 16:32:53 +09:00
+++ b/include/net/ip6_route.h 2004-09-13 16:32:53 +09:00
@@ -92,6 +92,7 @@
extern void rt6_redirect(struct in6_addr *dest,
struct in6_addr *saddr,
struct neighbour *neigh,
+ u8 *lladdr,
int on_link);
extern void rt6_pmtu_discovery(struct in6_addr *daddr,
diff -Nru a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
--- a/net/ipv6/ndisc.c 2004-09-13 16:32:53 +09:00
+++ b/net/ipv6/ndisc.c 2004-09-13 16:32:53 +09:00
@@ -1204,24 +1204,11 @@
return;
}
}
- /* passed validation tests */
-
- /*
- We install redirect only if nexthop state is valid.
- */
neigh = __neigh_lookup(&nd_tbl, target, skb->dev, 1);
if (neigh) {
- neigh_update(neigh, lladdr, NUD_STALE,
- NEIGH_UPDATE_F_WEAK_OVERRIDE|
- NEIGH_UPDATE_F_OVERRIDE|
- (on_link ? 0 : (NEIGH_UPDATE_F_OVERRIDE_ISROUTER|
- NEIGH_UPDATE_F_ISROUTER))
- );
- if (neigh->nud_state&NUD_VALID)
- rt6_redirect(dest, &skb->nh.ipv6h->saddr, neigh, on_link);
- else
- __neigh_event_send(neigh, NULL);
+ rt6_redirect(dest, &skb->nh.ipv6h->saddr, neigh, lladdr,
+ on_link);
neigh_release(neigh);
}
in6_dev_put(in6_dev);
diff -Nru a/net/ipv6/route.c b/net/ipv6/route.c
--- a/net/ipv6/route.c 2004-09-13 16:32:53 +09:00
+++ b/net/ipv6/route.c 2004-09-13 16:32:53 +09:00
@@ -1007,7 +1007,7 @@
* Handle redirects
*/
void rt6_redirect(struct in6_addr *dest, struct in6_addr *saddr,
- struct neighbour *neigh, int on_link)
+ struct neighbour *neigh, u8 *lladdr, int on_link)
{
struct rt6_info *rt, *nrt;
@@ -1020,22 +1020,13 @@
if (neigh->dev != rt->rt6i_dev)
goto out;
- /* Redirect received -> path was valid.
- Look, redirects are sent only in response to data packets,
- so that this nexthop apparently is reachable. --ANK
- */
- dst_confirm(&rt->u.dst);
-
- /* Duplicate redirect: silently ignore. */
- if (neigh == rt->u.dst.neighbour)
- goto out;
-
- /* Current route is on-link; redirect is always invalid.
-
- Seems, previous statement is not true. It could
- be node, which looks for us as on-link (f.e. proxy ndisc)
- But then router serving it might decide, that we should
- know truth 8)8) --ANK (980726).
+ /*
+ * Current route is on-link; redirect is always invalid.
+ *
+ * Seems, previous statement is not true. It could
+ * be node, which looks for us as on-link (f.e. proxy ndisc)
+ * But then router serving it might decide, that we should
+ * know truth 8)8) --ANK (980726).
*/
if (!(rt->rt6i_flags&RTF_GATEWAY))
goto out;
@@ -1047,7 +1038,6 @@
* is a bit fuzzy and one might need to check all default
* routers.
*/
-
if (ipv6_addr_cmp(saddr, &rt->rt6i_gateway)) {
if (rt->rt6i_flags & RTF_DEFAULT) {
struct rt6_info *rt1;
@@ -1075,6 +1065,24 @@
/*
* We have finally decided to accept it.
*/
+
+ neigh_update(neigh, lladdr, NUD_STALE,
+ NEIGH_UPDATE_F_WEAK_OVERRIDE|
+ NEIGH_UPDATE_F_OVERRIDE|
+ (on_link ? 0 : (NEIGH_UPDATE_F_OVERRIDE_ISROUTER|
+ NEIGH_UPDATE_F_ISROUTER))
+ );
+
+ /*
+ * Redirect received -> path was valid.
+ * Look, redirects are sent only in response to data packets,
+ * so that this nexthop apparently is reachable. --ANK
+ */
+ dst_confirm(&rt->u.dst);
+
+ /* Duplicate redirect: silently ignore. */
+ if (neigh == rt->u.dst.neighbour)
+ goto out;
nrt = ip6_rt_copy(rt);
if (nrt == NULL)
ChangeSet@1.1877, 2004-09-13 15:59:11+09:00, yoshfuji@linux-ipv6.org
[IPV6] don't use expired default routes.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/net/ipv6/route.c b/net/ipv6/route.c
--- a/net/ipv6/route.c 2004-09-13 16:32:56 +09:00
+++ b/net/ipv6/route.c 2004-09-13 16:32:56 +09:00
@@ -227,6 +227,10 @@
sprt->rt6i_dev->ifindex == oif))
m += 8;
+ if (sprt->rt6i_expires &&
+ time_after(jiffies, sprt->rt6i_expires))
+ continue;
+
if (sprt == rt6_dflt_pointer)
m += 4;
ChangeSet@1.1878, 2004-09-13 16:02:20+09:00, yoshfuji@linux-ipv6.org
[IPV6] ensure to aging default routes.
This patch is product of corraboration with Ville Nuorvala
<vnuorval@tcs.hut.fi>.
Signed-off-by: Hideaki YOSHIFUJi <yoshfuji@linux-ipv6.org>
diff -Nru a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
--- a/net/ipv6/ip6_fib.c 2004-09-13 16:32:59 +09:00
+++ b/net/ipv6/ip6_fib.c 2004-09-13 16:32:59 +09:00
@@ -49,6 +49,9 @@
struct rt6_statistics rt6_stats;
+extern struct rt6_info *rt6_dflt_pointer;
+extern spinlock_t rt6_dflt_lock;
+
static kmem_cache_t * fib6_node_kmem;
enum fib_walk_state_t
@@ -1184,6 +1187,10 @@
if (rt->rt6i_flags&RTF_EXPIRES && rt->rt6i_expires) {
if (time_after(now, rt->rt6i_expires)) {
RT6_TRACE("expiring %p\n", rt);
+ spin_lock_bh(&rt6_dflt_lock);
+ if (rt == rt6_dflt_pointer)
+ rt6_dflt_pointer = NULL;
+ spin_unlock_bh(&rt6_dflt_lock);
return -1;
}
gc_args.more++;
diff -Nru a/net/ipv6/route.c b/net/ipv6/route.c
--- a/net/ipv6/route.c 2004-09-13 16:32:59 +09:00
+++ b/net/ipv6/route.c 2004-09-13 16:32:59 +09:00
@@ -208,8 +208,8 @@
/*
* pointer to the last default router chosen. BH is disabled locally.
*/
-static struct rt6_info *rt6_dflt_pointer;
-static spinlock_t rt6_dflt_lock = SPIN_LOCK_UNLOCKED;
+struct rt6_info *rt6_dflt_pointer;
+spinlock_t rt6_dflt_lock = SPIN_LOCK_UNLOCKED;
/* Default Router Selection (RFC 2461 6.3.6) */
static struct rt6_info *rt6_best_dflt(struct rt6_info *rt, int oif)
@@ -227,7 +227,7 @@
sprt->rt6i_dev->ifindex == oif))
m += 8;
- if (sprt->rt6i_expires &&
+ if ((sprt->rt6i_flags & RTF_EXPIRES) &&
time_after(jiffies, sprt->rt6i_expires))
continue;
@@ -1265,7 +1265,7 @@
rtmsg.rtmsg_type = RTMSG_NEWROUTE;
ipv6_addr_copy(&rtmsg.rtmsg_gateway, gwaddr);
rtmsg.rtmsg_metric = 1024;
- rtmsg.rtmsg_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT | RTF_UP;
+ rtmsg.rtmsg_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT | RTF_UP | RTF_EXPIRES;
rtmsg.rtmsg_ifindex = dev->ifindex;
ChangeSet@1.1879, 2004-09-13 16:04:16+09:00, yoshfuji@linux-ipv6.org
[IPV6] purge routes via non-router neighbour but gateway.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
diff -Nru a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
--- a/net/ipv6/ip6_fib.c 2004-09-13 16:33:02 +09:00
+++ b/net/ipv6/ip6_fib.c 2004-09-13 16:33:02 +09:00
@@ -1199,6 +1199,11 @@
time_after_eq(now, rt->u.dst.lastuse + gc_args.timeout)) {
RT6_TRACE("aging clone %p\n", rt);
return -1;
+ } else if ((rt->rt6i_flags & RTF_GATEWAY) &&
+ (!(rt->rt6i_nexthop->flags & NTF_ROUTER))) {
+ RT6_TRACE("purging route %p via non-router but gateway\n",
+ rt);
+ return -1;
}
gc_args.more++;
}
--
Hideaki YOSHIFUJI @ USAGI Project <yoshfuji@linux-ipv6.org>
GPG FP: 9022 65EB 1ECF 3AD1 0BDF 80D8 4807 F894 E062 0EEA
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [BK PATCH] [IPV6] Merge Specification Conformity Improvements
2004-09-13 14:17 [BK PATCH] [IPV6] Merge Specification Conformity Improvements YOSHIFUJI Hideaki / 吉藤英明
@ 2004-09-13 14:29 ` Pekka Savola
2004-09-13 15:01 ` YOSHIFUJI Hideaki / 吉藤英明
2004-09-13 20:49 ` David S. Miller
2004-09-13 22:55 ` David S. Miller
2 siblings, 1 reply; 6+ messages in thread
From: Pekka Savola @ 2004-09-13 14:29 UTC (permalink / raw)
To: YOSHIFUJI Hideaki / 吉藤英明
Cc: davem, netdev, vnuorval
Thanks for doing these; I hope you guys will have energy for the other
spec fixes to come :)
One thing I noted when reading the comment:
On Mon, 13 Sep 2004, YOSHIFUJI Hideaki / [iso-2022-jp] ^[$B5HF#1QL@^[(B wrote:
> + /*
> + * Redirect received -> path was valid.
> + * Look, redirects are sent only in response to data packets,
> + * so that this nexthop apparently is reachable. --ANK
> + */
> + dst_confirm(&rt->u.dst);
> +
> + /* Duplicate redirect: silently ignore. */
> + if (neigh == rt->u.dst.neighbour)
> + goto out;
The above applies for "valid" redirects, which have been received
based on the traffic sent.
However, if someone would be forging redirects, the comment would no
longer hold.
I don't know the implications in this case: whether the code needs to
have different assumptions wrt. source of redirects, or whether this
is just a wording issue in the comment above.
--
Pekka Savola "You each name yourselves king, yet the
Netcore Oy kingdom bleeds."
Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [BK PATCH] [IPV6] Merge Specification Conformity Improvements
2004-09-13 14:29 ` Pekka Savola
@ 2004-09-13 15:01 ` YOSHIFUJI Hideaki / 吉藤英明
0 siblings, 0 replies; 6+ messages in thread
From: YOSHIFUJI Hideaki / 吉藤英明 @ 2004-09-13 15:01 UTC (permalink / raw)
To: pekkas; +Cc: davem, netdev, vnuorval, yoshfuji
In article <Pine.LNX.4.44.0409131725250.22572-100000@netcore.fi> (at Mon, 13 Sep 2004 17:29:20 +0300 (EEST)), Pekka Savola <pekkas@netcore.fi> says:
> However, if someone would be forging redirects, the comment would no
> longer hold.
>
> I don't know the implications in this case: whether the code needs to
> have different assumptions wrt. source of redirects, or whether this
> is just a wording issue in the comment above.
I think we're protected (at least) as the standards says.
eg.
- off-link attacks
- redirect from non-router for the destination
etc.
I'm not sure if we have other things we can do against this issue.
I think other on-link issues (including "forging" issues)
will be solved by SEND (SEcuring Neighbor Discovery).
--yoshfuji
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [BK PATCH] [IPV6] Merge Specification Conformity Improvements
2004-09-13 14:17 [BK PATCH] [IPV6] Merge Specification Conformity Improvements YOSHIFUJI Hideaki / 吉藤英明
2004-09-13 14:29 ` Pekka Savola
@ 2004-09-13 20:49 ` David S. Miller
2004-09-13 22:55 ` David S. Miller
2 siblings, 0 replies; 6+ messages in thread
From: David S. Miller @ 2004-09-13 20:49 UTC (permalink / raw)
To: yoshfuji; +Cc: netdev, vnuorval
On Mon, 13 Sep 2004 23:17:32 +0900 (JST)
YOSHIFUJI Hideaki / ^[$B5HF#1QL@^[(B <yoshfuji@linux-ipv6.org> wrote:
> Here're the changesets which improve conformity to
> the IPv6 specifications / standards.
>
> Please pull from
> <bk://bk.skbuff.net:20609/linux-2.6-ndp-20040913/>
I am pulling this, thank you.
There are at least two issues I wish to discuss and
that will show up in a follow mail later today after
I read these changes some more.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [BK PATCH] [IPV6] Merge Specification Conformity Improvements
2004-09-13 14:17 [BK PATCH] [IPV6] Merge Specification Conformity Improvements YOSHIFUJI Hideaki / 吉藤英明
2004-09-13 14:29 ` Pekka Savola
2004-09-13 20:49 ` David S. Miller
@ 2004-09-13 22:55 ` David S. Miller
2004-09-14 1:19 ` YOSHIFUJI Hideaki / 吉藤英明
2 siblings, 1 reply; 6+ messages in thread
From: David S. Miller @ 2004-09-13 22:55 UTC (permalink / raw)
To: yoshfuji; +Cc: netdev, vnuorval
Ok, here are comments :-)
1) wrt. net/core/neighbour.c changes to use time_after() and
similar interfaces. Here is why I didn't do that previously.
I showed this to Alexey some time ago and he made me aware of
some "black magic" in this area. Here is what he told me:
-------------------------------------------------------------
BTW while I remember... Long time ago (when no such macros even existed)
I definitely used some funny technique. Namely, when you definitely
know that some timestamp is growing, now - ts_prev > timeo
is always right thing and time_after(now, ts_prev+timeo) is not.
Unfortunately, I do not remember where exactly. It is sick approach, of course,
but it allowed not to switch to timeval to stamp long living objects.
So, if you will notice such place, it would be better to make it to use
timeval rather than to time_* macros.
-------------------------------------------------------------
2) rt6_dflt_{pointer,lock}
Maybe it would be better to export a function that operates
on these objects rather than the objects themselves. That
way we could keep them and their implementation static to
ip6_fib.c
These are minor issues though, and as I stated I pulled your changes
in already.
Thanks.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [BK PATCH] [IPV6] Merge Specification Conformity Improvements
2004-09-13 22:55 ` David S. Miller
@ 2004-09-14 1:19 ` YOSHIFUJI Hideaki / 吉藤英明
0 siblings, 0 replies; 6+ messages in thread
From: YOSHIFUJI Hideaki / 吉藤英明 @ 2004-09-14 1:19 UTC (permalink / raw)
To: davem; +Cc: netdev, vnuorval, yoshfuji
In article <20040913155506.4c99d4c8.davem@davemloft.net> (at Mon, 13 Sep 2004 15:55:06 -0700), "David S. Miller" <davem@davemloft.net> says:
> 2) rt6_dflt_{pointer,lock}
>
> Maybe it would be better to export a function that operates
> on these objects rather than the objects themselves. That
> way we could keep them and their implementation static to
> ip6_fib.c
Yes! I know it is silly, of course.
Well, I have a plan to remove rt6_dflt_{pointer,lock} shortly,
but anyway...
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
===== include/net/ip6_route.h 1.18 vs edited =====
--- 1.18/include/net/ip6_route.h 2004-09-11 23:50:37 +09:00
+++ edited/include/net/ip6_route.h 2004-09-14 10:01:50 +09:00
@@ -89,6 +89,8 @@
extern void rt6_purge_dflt_routers(int lst_resort);
+extern void rt6_reset_dflt_pointer(struct rt6_info *rt);
+
extern void rt6_redirect(struct in6_addr *dest,
struct in6_addr *saddr,
struct neighbour *neigh,
===== net/ipv6/ip6_fib.c 1.30 vs edited =====
--- 1.30/net/ipv6/ip6_fib.c 2004-09-11 23:56:06 +09:00
+++ edited/net/ipv6/ip6_fib.c 2004-09-14 10:00:33 +09:00
@@ -49,9 +49,6 @@
struct rt6_statistics rt6_stats;
-extern struct rt6_info *rt6_dflt_pointer;
-extern spinlock_t rt6_dflt_lock;
-
static kmem_cache_t * fib6_node_kmem;
enum fib_walk_state_t
@@ -1187,10 +1184,7 @@
if (rt->rt6i_flags&RTF_EXPIRES && rt->rt6i_expires) {
if (time_after(now, rt->rt6i_expires)) {
RT6_TRACE("expiring %p\n", rt);
- spin_lock_bh(&rt6_dflt_lock);
- if (rt == rt6_dflt_pointer)
- rt6_dflt_pointer = NULL;
- spin_unlock_bh(&rt6_dflt_lock);
+ rt6_reset_dflt_pointer(rt);
return -1;
}
gc_args.more++;
===== net/ipv6/route.c 1.92 vs edited =====
--- 1.92/net/ipv6/route.c 2004-09-11 23:53:53 +09:00
+++ edited/net/ipv6/route.c 2004-09-14 10:10:25 +09:00
@@ -210,6 +210,16 @@
struct rt6_info *rt6_dflt_pointer;
spinlock_t rt6_dflt_lock = SPIN_LOCK_UNLOCKED;
+void rt6_reset_dflt_pointer(struct rt6_info *rt)
+{
+ spin_lock_bh(&rt6_dflt_lock);
+ if (rt == NULL || rt == rt6_dflt_pointer) {
+ RT6_TRACE("reset default router: %p->NULL\n", rt6_dflt_pointer);
+ rt6_dflt_pointer = NULL;
+ }
+ spin_unlock_bh(&rt6_dflt_lock);
+}
+
/* Default Router Selection (RFC 2461 6.3.6) */
static struct rt6_info *rt6_best_dflt(struct rt6_info *rt, int oif)
{
@@ -959,9 +969,7 @@
write_lock_bh(&rt6_lock);
- spin_lock_bh(&rt6_dflt_lock);
- rt6_dflt_pointer = NULL;
- spin_unlock_bh(&rt6_dflt_lock);
+ rt6_reset_dflt_pointer(NULL);
dst_release(&rt->u.dst);
@@ -1288,9 +1296,7 @@
if (rt->rt6i_flags & flags) {
dst_hold(&rt->u.dst);
- spin_lock_bh(&rt6_dflt_lock);
- rt6_dflt_pointer = NULL;
- spin_unlock_bh(&rt6_dflt_lock);
+ rt6_reset_dflt_pointer(NULL);
read_unlock_bh(&rt6_lock);
--
Hideaki YOSHIFUJI @ USAGI Project <yoshfuji@linux-ipv6.org>
GPG FP: 9022 65EB 1ECF 3AD1 0BDF 80D8 4807 F894 E062 0EEA
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2004-09-14 1:19 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-09-13 14:17 [BK PATCH] [IPV6] Merge Specification Conformity Improvements YOSHIFUJI Hideaki / 吉藤英明
2004-09-13 14:29 ` Pekka Savola
2004-09-13 15:01 ` YOSHIFUJI Hideaki / 吉藤英明
2004-09-13 20:49 ` David S. Miller
2004-09-13 22:55 ` David S. Miller
2004-09-14 1:19 ` YOSHIFUJI Hideaki / 吉藤英明
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).