2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Linux INET6 implementation
|
|
|
|
* FIB front-end.
|
|
|
|
*
|
|
|
|
* Authors:
|
2007-02-09 07:24:49 -07:00
|
|
|
* Pedro Roque <roque@di.fc.ul.pt>
|
2005-04-16 16:20:36 -06:00
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public License
|
|
|
|
* as published by the Free Software Foundation; either version
|
|
|
|
* 2 of the License, or (at your option) any later version.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* Changes:
|
|
|
|
*
|
|
|
|
* YOSHIFUJI Hideaki @USAGI
|
|
|
|
* reworked default router selection.
|
|
|
|
* - respect outgoing interface
|
|
|
|
* - select from (probably) reachable routers (i.e.
|
|
|
|
* routers in REACHABLE, STALE, DELAY or PROBE states).
|
|
|
|
* - always select the same router if it is (probably)
|
|
|
|
* reachable. otherwise, round-robin the list.
|
2006-08-23 18:23:25 -06:00
|
|
|
* Ville Nuorvala
|
|
|
|
* Fixed routing subtrees.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
|
2012-05-15 08:11:53 -06:00
|
|
|
#define pr_fmt(fmt) "IPv6: " fmt
|
|
|
|
|
2006-01-11 13:17:47 -07:00
|
|
|
#include <linux/capability.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/errno.h>
|
2011-07-15 09:47:34 -06:00
|
|
|
#include <linux/export.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/times.h>
|
|
|
|
#include <linux/socket.h>
|
|
|
|
#include <linux/sockios.h>
|
|
|
|
#include <linux/net.h>
|
|
|
|
#include <linux/route.h>
|
|
|
|
#include <linux/netdevice.h>
|
|
|
|
#include <linux/in6.h>
|
2008-04-02 18:22:53 -06:00
|
|
|
#include <linux/mroute6.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/if_arp.h>
|
|
|
|
#include <linux/proc_fs.h>
|
|
|
|
#include <linux/seq_file.h>
|
2008-03-04 00:28:58 -07:00
|
|
|
#include <linux/nsproxy.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 02:04:11 -06:00
|
|
|
#include <linux/slab.h>
|
2007-09-12 04:01:34 -06:00
|
|
|
#include <net/net_namespace.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <net/snmp.h>
|
|
|
|
#include <net/ipv6.h>
|
|
|
|
#include <net/ip6_fib.h>
|
|
|
|
#include <net/ip6_route.h>
|
|
|
|
#include <net/ndisc.h>
|
|
|
|
#include <net/addrconf.h>
|
|
|
|
#include <net/tcp.h>
|
|
|
|
#include <linux/rtnetlink.h>
|
|
|
|
#include <net/dst.h>
|
|
|
|
#include <net/xfrm.h>
|
2006-07-30 21:43:36 -06:00
|
|
|
#include <net/netevent.h>
|
2006-08-15 01:35:24 -06:00
|
|
|
#include <net/netlink.h>
|
2012-10-21 21:42:09 -06:00
|
|
|
#include <net/nexthop.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#include <asm/uaccess.h>
|
|
|
|
|
|
|
|
#ifdef CONFIG_SYSCTL
|
|
|
|
#include <linux/sysctl.h>
|
|
|
|
#endif
|
|
|
|
|
2013-07-11 04:43:42 -06:00
|
|
|
enum rt6_nud_state {
|
|
|
|
RT6_NUD_FAIL_HARD = -2,
|
|
|
|
RT6_NUD_FAIL_SOFT = -1,
|
|
|
|
RT6_NUD_SUCCEED = 1
|
|
|
|
};
|
|
|
|
|
2012-04-05 18:13:10 -06:00
|
|
|
static struct rt6_info *ip6_rt_copy(struct rt6_info *ort,
|
2011-07-19 14:18:36 -06:00
|
|
|
const struct in6_addr *dest);
|
2005-04-16 16:20:36 -06:00
|
|
|
static struct dst_entry *ip6_dst_check(struct dst_entry *dst, u32 cookie);
|
2010-12-13 13:52:14 -07:00
|
|
|
static unsigned int ip6_default_advmss(const struct dst_entry *dst);
|
2011-11-22 19:12:51 -07:00
|
|
|
static unsigned int ip6_mtu(const struct dst_entry *dst);
|
2005-04-16 16:20:36 -06:00
|
|
|
static struct dst_entry *ip6_negative_advice(struct dst_entry *);
|
|
|
|
static void ip6_dst_destroy(struct dst_entry *);
|
|
|
|
static void ip6_dst_ifdown(struct dst_entry *,
|
|
|
|
struct net_device *dev, int how);
|
2008-01-18 04:56:57 -07:00
|
|
|
static int ip6_dst_gc(struct dst_ops *ops);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
static int ip6_pkt_discard(struct sk_buff *skb);
|
|
|
|
static int ip6_pkt_discard_out(struct sk_buff *skb);
|
2013-12-02 07:25:21 -07:00
|
|
|
static int ip6_pkt_prohibit(struct sk_buff *skb);
|
|
|
|
static int ip6_pkt_prohibit_out(struct sk_buff *skb);
|
2005-04-16 16:20:36 -06:00
|
|
|
static void ip6_link_failure(struct sk_buff *skb);
|
2012-07-17 04:29:28 -06:00
|
|
|
static void ip6_rt_update_pmtu(struct dst_entry *dst, struct sock *sk,
|
|
|
|
struct sk_buff *skb, u32 mtu);
|
|
|
|
static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk,
|
|
|
|
struct sk_buff *skb);
|
2013-06-28 09:35:48 -06:00
|
|
|
static int rt6_score_route(struct rt6_info *rt, int oif, int strict);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-03-20 18:06:24 -07:00
|
|
|
#ifdef CONFIG_IPV6_ROUTE_INFO
|
2008-03-04 14:46:48 -07:00
|
|
|
static struct rt6_info *rt6_add_route_info(struct net *net,
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct in6_addr *prefix, int prefixlen,
|
|
|
|
const struct in6_addr *gwaddr, int ifindex,
|
2012-04-14 23:58:06 -06:00
|
|
|
unsigned int pref);
|
2008-03-04 14:46:48 -07:00
|
|
|
static struct rt6_info *rt6_get_route_info(struct net *net,
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct in6_addr *prefix, int prefixlen,
|
|
|
|
const struct in6_addr *gwaddr, int ifindex);
|
2006-03-20 18:06:24 -07:00
|
|
|
#endif
|
|
|
|
|
2011-01-27 15:58:42 -07:00
|
|
|
static u32 *ipv6_cow_metrics(struct dst_entry *dst, unsigned long old)
|
|
|
|
{
|
|
|
|
struct rt6_info *rt = (struct rt6_info *) dst;
|
|
|
|
struct inet_peer *peer;
|
|
|
|
u32 *p = NULL;
|
|
|
|
|
2011-09-05 15:34:30 -06:00
|
|
|
if (!(rt->dst.flags & DST_HOST))
|
|
|
|
return NULL;
|
|
|
|
|
2012-06-09 00:24:18 -06:00
|
|
|
peer = rt6_get_peer_create(rt);
|
2011-01-27 15:58:42 -07:00
|
|
|
if (peer) {
|
|
|
|
u32 *old_p = __DST_METRICS_PTR(old);
|
|
|
|
unsigned long prev, new;
|
|
|
|
|
|
|
|
p = peer->metrics;
|
|
|
|
if (inet_metrics_new(peer))
|
|
|
|
memcpy(p, old_p, sizeof(u32) * RTAX_MAX);
|
|
|
|
|
|
|
|
new = (unsigned long) p;
|
|
|
|
prev = cmpxchg(&dst->_metrics, old, new);
|
|
|
|
|
|
|
|
if (prev != old) {
|
|
|
|
p = __DST_METRICS_PTR(prev);
|
|
|
|
if (prev & DST_METRICS_READ_ONLY)
|
|
|
|
p = NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return p;
|
|
|
|
}
|
|
|
|
|
2012-07-02 22:52:24 -06:00
|
|
|
static inline const void *choose_neigh_daddr(struct rt6_info *rt,
|
|
|
|
struct sk_buff *skb,
|
|
|
|
const void *daddr)
|
2012-01-26 13:22:32 -07:00
|
|
|
{
|
|
|
|
struct in6_addr *p = &rt->rt6i_gateway;
|
|
|
|
|
2012-01-26 14:29:16 -07:00
|
|
|
if (!ipv6_addr_any(p))
|
2012-01-26 13:22:32 -07:00
|
|
|
return (const void *) p;
|
2012-07-02 22:52:24 -06:00
|
|
|
else if (skb)
|
|
|
|
return &ipv6_hdr(skb)->daddr;
|
2012-01-26 13:22:32 -07:00
|
|
|
return daddr;
|
|
|
|
}
|
|
|
|
|
2012-07-02 22:52:24 -06:00
|
|
|
static struct neighbour *ip6_neigh_lookup(const struct dst_entry *dst,
|
|
|
|
struct sk_buff *skb,
|
|
|
|
const void *daddr)
|
2011-07-18 01:40:17 -06:00
|
|
|
{
|
2012-01-26 13:22:32 -07:00
|
|
|
struct rt6_info *rt = (struct rt6_info *) dst;
|
|
|
|
struct neighbour *n;
|
|
|
|
|
2012-07-02 22:52:24 -06:00
|
|
|
daddr = choose_neigh_daddr(rt, skb, daddr);
|
2013-01-17 05:53:09 -07:00
|
|
|
n = __ipv6_neigh_lookup(dst->dev, daddr);
|
2011-12-28 13:41:23 -07:00
|
|
|
if (n)
|
|
|
|
return n;
|
|
|
|
return neigh_create(&nd_tbl, daddr, dst->dev);
|
|
|
|
}
|
|
|
|
|
2008-03-04 14:48:53 -07:00
|
|
|
static struct dst_ops ip6_dst_ops_template = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.family = AF_INET6,
|
2009-02-01 01:45:17 -07:00
|
|
|
.protocol = cpu_to_be16(ETH_P_IPV6),
|
2005-04-16 16:20:36 -06:00
|
|
|
.gc = ip6_dst_gc,
|
|
|
|
.gc_thresh = 1024,
|
|
|
|
.check = ip6_dst_check,
|
2010-12-13 13:52:14 -07:00
|
|
|
.default_advmss = ip6_default_advmss,
|
2011-11-22 19:12:51 -07:00
|
|
|
.mtu = ip6_mtu,
|
2011-01-27 15:58:42 -07:00
|
|
|
.cow_metrics = ipv6_cow_metrics,
|
2005-04-16 16:20:36 -06:00
|
|
|
.destroy = ip6_dst_destroy,
|
|
|
|
.ifdown = ip6_dst_ifdown,
|
|
|
|
.negative_advice = ip6_negative_advice,
|
|
|
|
.link_failure = ip6_link_failure,
|
|
|
|
.update_pmtu = ip6_rt_update_pmtu,
|
2012-07-12 01:05:02 -06:00
|
|
|
.redirect = rt6_do_redirect,
|
2008-05-20 15:32:14 -06:00
|
|
|
.local_out = __ip6_local_out,
|
2011-07-18 01:40:17 -06:00
|
|
|
.neigh_lookup = ip6_neigh_lookup,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2011-11-22 19:12:51 -07:00
|
|
|
static unsigned int ip6_blackhole_mtu(const struct dst_entry *dst)
|
2011-01-31 14:16:00 -07:00
|
|
|
{
|
2011-11-22 19:13:31 -07:00
|
|
|
unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
|
|
|
|
|
|
|
|
return mtu ? : dst->dev->mtu;
|
2011-01-31 14:16:00 -07:00
|
|
|
}
|
|
|
|
|
2012-07-17 04:29:28 -06:00
|
|
|
static void ip6_rt_blackhole_update_pmtu(struct dst_entry *dst, struct sock *sk,
|
|
|
|
struct sk_buff *skb, u32 mtu)
|
2007-05-24 19:17:54 -06:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2012-07-17 04:29:28 -06:00
|
|
|
static void ip6_rt_blackhole_redirect(struct dst_entry *dst, struct sock *sk,
|
|
|
|
struct sk_buff *skb)
|
2012-07-12 01:39:24 -06:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2011-04-24 16:07:32 -06:00
|
|
|
static u32 *ip6_rt_blackhole_cow_metrics(struct dst_entry *dst,
|
|
|
|
unsigned long old)
|
|
|
|
{
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2007-05-24 19:17:54 -06:00
|
|
|
static struct dst_ops ip6_dst_blackhole_ops = {
|
|
|
|
.family = AF_INET6,
|
2009-02-01 01:45:17 -07:00
|
|
|
.protocol = cpu_to_be16(ETH_P_IPV6),
|
2007-05-24 19:17:54 -06:00
|
|
|
.destroy = ip6_dst_destroy,
|
|
|
|
.check = ip6_dst_check,
|
2011-11-22 19:12:51 -07:00
|
|
|
.mtu = ip6_blackhole_mtu,
|
2011-02-18 12:39:01 -07:00
|
|
|
.default_advmss = ip6_default_advmss,
|
2007-05-24 19:17:54 -06:00
|
|
|
.update_pmtu = ip6_rt_blackhole_update_pmtu,
|
2012-07-12 01:39:24 -06:00
|
|
|
.redirect = ip6_rt_blackhole_redirect,
|
2011-04-24 16:07:32 -06:00
|
|
|
.cow_metrics = ip6_rt_blackhole_cow_metrics,
|
2011-07-18 01:40:17 -06:00
|
|
|
.neigh_lookup = ip6_neigh_lookup,
|
2007-05-24 19:17:54 -06:00
|
|
|
};
|
|
|
|
|
net: Implement read-only protection and COW'ing of metrics.
Routing metrics are now copy-on-write.
Initially a route entry points it's metrics at a read-only location.
If a routing table entry exists, it will point there. Else it will
point at the all zero metric place-holder called 'dst_default_metrics'.
The writeability state of the metrics is stored in the low bits of the
metrics pointer, we have two bits left to spare if we want to store
more states.
For the initial implementation, COW is implemented simply via kmalloc.
However future enhancements will change this to place the writable
metrics somewhere else, in order to increase sharing. Very likely
this "somewhere else" will be the inetpeer cache.
Note also that this means that metrics updates may transiently fail
if we cannot COW the metrics successfully.
But even by itself, this patch should decrease memory usage and
increase cache locality especially for routing workloads. In those
cases the read-only metric copies stay in place and never get written
to.
TCP workloads where metrics get updated, and those rare cases where
PMTU triggers occur, will take a very slight performance hit. But
that hit will be alleviated when the long-term writable metrics
move to a more sharable location.
Since the metrics storage went from a u32 array of RTAX_MAX entries to
what is essentially a pointer, some retooling of the dst_entry layout
was necessary.
Most importantly, we need to preserve the alignment of the reference
count so that it doesn't share cache lines with the read-mostly state,
as per Eric Dumazet's alignment assertion checks.
The only non-trivial bit here is the move of the 'flags' member into
the writeable cacheline. This is OK since we are always accessing the
flags around the same moment when we made a modification to the
reference count.
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-26 21:51:05 -07:00
|
|
|
static const u32 ip6_template_metrics[RTAX_MAX] = {
|
2012-10-24 00:01:18 -06:00
|
|
|
[RTAX_HOPLIMIT - 1] = 0,
|
net: Implement read-only protection and COW'ing of metrics.
Routing metrics are now copy-on-write.
Initially a route entry points it's metrics at a read-only location.
If a routing table entry exists, it will point there. Else it will
point at the all zero metric place-holder called 'dst_default_metrics'.
The writeability state of the metrics is stored in the low bits of the
metrics pointer, we have two bits left to spare if we want to store
more states.
For the initial implementation, COW is implemented simply via kmalloc.
However future enhancements will change this to place the writable
metrics somewhere else, in order to increase sharing. Very likely
this "somewhere else" will be the inetpeer cache.
Note also that this means that metrics updates may transiently fail
if we cannot COW the metrics successfully.
But even by itself, this patch should decrease memory usage and
increase cache locality especially for routing workloads. In those
cases the read-only metric copies stay in place and never get written
to.
TCP workloads where metrics get updated, and those rare cases where
PMTU triggers occur, will take a very slight performance hit. But
that hit will be alleviated when the long-term writable metrics
move to a more sharable location.
Since the metrics storage went from a u32 array of RTAX_MAX entries to
what is essentially a pointer, some retooling of the dst_entry layout
was necessary.
Most importantly, we need to preserve the alignment of the reference
count so that it doesn't share cache lines with the read-mostly state,
as per Eric Dumazet's alignment assertion checks.
The only non-trivial bit here is the move of the 'flags' member into
the writeable cacheline. This is OK since we are always accessing the
flags around the same moment when we made a modification to the
reference count.
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-26 21:51:05 -07:00
|
|
|
};
|
|
|
|
|
2012-09-11 15:47:51 -06:00
|
|
|
static const struct rt6_info ip6_null_entry_template = {
|
2010-06-11 00:31:35 -06:00
|
|
|
.dst = {
|
|
|
|
.__refcnt = ATOMIC_INIT(1),
|
|
|
|
.__use = 1,
|
2012-09-10 16:09:47 -06:00
|
|
|
.obsolete = DST_OBSOLETE_FORCE_CHK,
|
2010-06-11 00:31:35 -06:00
|
|
|
.error = -ENETUNREACH,
|
|
|
|
.input = ip6_pkt_discard,
|
|
|
|
.output = ip6_pkt_discard_out,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
.rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP),
|
2009-05-20 18:38:59 -06:00
|
|
|
.rt6i_protocol = RTPROT_KERNEL,
|
2005-04-16 16:20:36 -06:00
|
|
|
.rt6i_metric = ~(u32) 0,
|
|
|
|
.rt6i_ref = ATOMIC_INIT(1),
|
|
|
|
};
|
|
|
|
|
2006-08-04 04:39:02 -06:00
|
|
|
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
|
|
|
|
|
2012-09-11 15:47:51 -06:00
|
|
|
static const struct rt6_info ip6_prohibit_entry_template = {
|
2010-06-11 00:31:35 -06:00
|
|
|
.dst = {
|
|
|
|
.__refcnt = ATOMIC_INIT(1),
|
|
|
|
.__use = 1,
|
2012-09-10 16:09:47 -06:00
|
|
|
.obsolete = DST_OBSOLETE_FORCE_CHK,
|
2010-06-11 00:31:35 -06:00
|
|
|
.error = -EACCES,
|
|
|
|
.input = ip6_pkt_prohibit,
|
|
|
|
.output = ip6_pkt_prohibit_out,
|
2006-08-04 04:39:02 -06:00
|
|
|
},
|
|
|
|
.rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP),
|
2009-05-20 18:38:59 -06:00
|
|
|
.rt6i_protocol = RTPROT_KERNEL,
|
2006-08-04 04:39:02 -06:00
|
|
|
.rt6i_metric = ~(u32) 0,
|
|
|
|
.rt6i_ref = ATOMIC_INIT(1),
|
|
|
|
};
|
|
|
|
|
2012-09-11 15:47:51 -06:00
|
|
|
static const struct rt6_info ip6_blk_hole_entry_template = {
|
2010-06-11 00:31:35 -06:00
|
|
|
.dst = {
|
|
|
|
.__refcnt = ATOMIC_INIT(1),
|
|
|
|
.__use = 1,
|
2012-09-10 16:09:47 -06:00
|
|
|
.obsolete = DST_OBSOLETE_FORCE_CHK,
|
2010-06-11 00:31:35 -06:00
|
|
|
.error = -EINVAL,
|
|
|
|
.input = dst_discard,
|
|
|
|
.output = dst_discard,
|
2006-08-04 04:39:02 -06:00
|
|
|
},
|
|
|
|
.rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP),
|
2009-05-20 18:38:59 -06:00
|
|
|
.rt6i_protocol = RTPROT_KERNEL,
|
2006-08-04 04:39:02 -06:00
|
|
|
.rt6i_metric = ~(u32) 0,
|
|
|
|
.rt6i_ref = ATOMIC_INIT(1),
|
|
|
|
};
|
|
|
|
|
|
|
|
#endif
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/* allocate dst with ip6_dst_ops */
|
2012-06-09 23:36:36 -06:00
|
|
|
static inline struct rt6_info *ip6_dst_alloc(struct net *net,
|
2011-06-24 16:25:00 -06:00
|
|
|
struct net_device *dev,
|
2012-06-11 03:01:56 -06:00
|
|
|
int flags,
|
|
|
|
struct fib6_table *table)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2012-06-09 23:36:36 -06:00
|
|
|
struct rt6_info *rt = dst_alloc(&net->ipv6.ip6_dst_ops, dev,
|
2012-09-10 16:09:46 -06:00
|
|
|
0, DST_OBSOLETE_FORCE_CHK, flags);
|
2011-04-28 15:31:47 -06:00
|
|
|
|
2012-06-09 23:36:36 -06:00
|
|
|
if (rt) {
|
2012-07-05 17:37:09 -06:00
|
|
|
struct dst_entry *dst = &rt->dst;
|
|
|
|
|
|
|
|
memset(dst + 1, 0, sizeof(*rt) - sizeof(*dst));
|
2012-06-11 03:01:56 -06:00
|
|
|
rt6_init_peer(rt, table ? &table->tb6_peers : net->ipv6.peers);
|
2013-07-29 18:33:53 -06:00
|
|
|
rt->rt6i_genid = rt_genid_ipv6(net);
|
2012-10-21 21:42:09 -06:00
|
|
|
INIT_LIST_HEAD(&rt->rt6i_siblings);
|
2012-06-09 23:36:36 -06:00
|
|
|
}
|
2011-04-28 15:31:47 -06:00
|
|
|
return rt;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static void ip6_dst_destroy(struct dst_entry *dst)
|
|
|
|
{
|
|
|
|
struct rt6_info *rt = (struct rt6_info *)dst;
|
|
|
|
struct inet6_dev *idev = rt->rt6i_idev;
|
ipv6: fix race condition regarding dst->expires and dst->from.
Eric Dumazet wrote:
| Some strange crashes happen in rt6_check_expired(), with access
| to random addresses.
|
| At first glance, it looks like the RTF_EXPIRES and
| stuff added in commit 1716a96101c49186b
| (ipv6: fix problem with expired dst cache)
| are racy : same dst could be manipulated at the same time
| on different cpus.
|
| At some point, our stack believes rt->dst.from contains a dst pointer,
| while its really a jiffie value (as rt->dst.expires shares the same area
| of memory)
|
| rt6_update_expires() should be fixed, or am I missing something ?
|
| CC Neil because of https://bugzilla.redhat.com/show_bug.cgi?id=892060
Because we do not have any locks for dst_entry, we cannot change
essential structure in the entry; e.g., we cannot change reference
to other entity.
To fix this issue, split 'from' and 'expires' field in dst_entry
out of union. Once it is 'from' is assigned in the constructor,
keep the reference until the very last stage of the life time of
the object.
Of course, it is unsafe to change 'from', so make rt6_set_from simple
just for fresh entries.
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Reported-by: Neil Horman <nhorman@tuxdriver.com>
CC: Gao Feng <gaofeng@cn.fujitsu.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reported-by: Steinar H. Gunderson <sesse@google.com>
Reviewed-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-02-19 17:29:08 -07:00
|
|
|
struct dst_entry *from = dst->from;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-09-05 15:34:30 -06:00
|
|
|
if (!(rt->dst.flags & DST_HOST))
|
|
|
|
dst_destroy_metrics_generic(dst);
|
|
|
|
|
2011-12-03 16:02:47 -07:00
|
|
|
if (idev) {
|
2005-04-16 16:20:36 -06:00
|
|
|
rt->rt6i_idev = NULL;
|
|
|
|
in6_dev_put(idev);
|
2007-02-09 07:24:49 -07:00
|
|
|
}
|
2012-04-05 18:13:10 -06:00
|
|
|
|
ipv6: fix race condition regarding dst->expires and dst->from.
Eric Dumazet wrote:
| Some strange crashes happen in rt6_check_expired(), with access
| to random addresses.
|
| At first glance, it looks like the RTF_EXPIRES and
| stuff added in commit 1716a96101c49186b
| (ipv6: fix problem with expired dst cache)
| are racy : same dst could be manipulated at the same time
| on different cpus.
|
| At some point, our stack believes rt->dst.from contains a dst pointer,
| while its really a jiffie value (as rt->dst.expires shares the same area
| of memory)
|
| rt6_update_expires() should be fixed, or am I missing something ?
|
| CC Neil because of https://bugzilla.redhat.com/show_bug.cgi?id=892060
Because we do not have any locks for dst_entry, we cannot change
essential structure in the entry; e.g., we cannot change reference
to other entity.
To fix this issue, split 'from' and 'expires' field in dst_entry
out of union. Once it is 'from' is assigned in the constructor,
keep the reference until the very last stage of the life time of
the object.
Of course, it is unsafe to change 'from', so make rt6_set_from simple
just for fresh entries.
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Reported-by: Neil Horman <nhorman@tuxdriver.com>
CC: Gao Feng <gaofeng@cn.fujitsu.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reported-by: Steinar H. Gunderson <sesse@google.com>
Reviewed-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-02-19 17:29:08 -07:00
|
|
|
dst->from = NULL;
|
|
|
|
dst_release(from);
|
2012-04-05 18:13:10 -06:00
|
|
|
|
2012-06-09 23:36:36 -06:00
|
|
|
if (rt6_has_peer(rt)) {
|
|
|
|
struct inet_peer *peer = rt6_peer_ptr(rt);
|
2010-11-30 13:27:11 -07:00
|
|
|
inet_putpeer(peer);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void rt6_bind_peer(struct rt6_info *rt, int create)
|
|
|
|
{
|
2012-06-09 23:36:36 -06:00
|
|
|
struct inet_peer_base *base;
|
2010-11-30 13:27:11 -07:00
|
|
|
struct inet_peer *peer;
|
|
|
|
|
2012-06-09 23:36:36 -06:00
|
|
|
base = inetpeer_base_ptr(rt->_rt6i_peer);
|
|
|
|
if (!base)
|
|
|
|
return;
|
|
|
|
|
|
|
|
peer = inet_getpeer_v6(base, &rt->rt6i_dst.addr, create);
|
2012-06-11 05:13:57 -06:00
|
|
|
if (peer) {
|
|
|
|
if (!rt6_set_peer(rt, peer))
|
|
|
|
inet_putpeer(peer);
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
|
|
|
|
int how)
|
|
|
|
{
|
|
|
|
struct rt6_info *rt = (struct rt6_info *)dst;
|
|
|
|
struct inet6_dev *idev = rt->rt6i_idev;
|
2007-12-07 01:38:10 -07:00
|
|
|
struct net_device *loopback_dev =
|
2008-03-25 06:47:49 -06:00
|
|
|
dev_net(dev)->loopback_dev;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2012-07-02 23:43:47 -06:00
|
|
|
if (dev != loopback_dev) {
|
|
|
|
if (idev && idev->dev == dev) {
|
|
|
|
struct inet6_dev *loopback_idev =
|
|
|
|
in6_dev_get(loopback_dev);
|
|
|
|
if (loopback_idev) {
|
|
|
|
rt->rt6i_idev = loopback_idev;
|
|
|
|
in6_dev_put(idev);
|
|
|
|
}
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-05-18 12:57:34 -06:00
|
|
|
static bool rt6_check_expired(const struct rt6_info *rt)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2012-04-05 18:13:10 -06:00
|
|
|
if (rt->rt6i_flags & RTF_EXPIRES) {
|
|
|
|
if (time_after(jiffies, rt->dst.expires))
|
2012-05-18 12:57:34 -06:00
|
|
|
return true;
|
2012-04-05 18:13:10 -06:00
|
|
|
} else if (rt->dst.from) {
|
2012-09-13 13:54:57 -06:00
|
|
|
return rt6_check_expired((struct rt6_info *) rt->dst.from);
|
2012-04-05 18:13:10 -06:00
|
|
|
}
|
2012-05-18 12:57:34 -06:00
|
|
|
return false;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2012-05-18 12:57:34 -06:00
|
|
|
static bool rt6_need_strict(const struct in6_addr *daddr)
|
2006-08-05 00:20:06 -06:00
|
|
|
{
|
2010-09-22 14:43:57 -06:00
|
|
|
return ipv6_addr_type(daddr) &
|
|
|
|
(IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK);
|
2006-08-05 00:20:06 -06:00
|
|
|
}
|
|
|
|
|
2012-10-21 21:42:09 -06:00
|
|
|
/* Multipath route selection:
|
|
|
|
* Hash based function using packet header and flowlabel.
|
|
|
|
* Adapted from fib_info_hashfn()
|
|
|
|
*/
|
|
|
|
static int rt6_info_hash_nhsfn(unsigned int candidate_count,
|
|
|
|
const struct flowi6 *fl6)
|
|
|
|
{
|
|
|
|
unsigned int val = fl6->flowi6_proto;
|
|
|
|
|
2013-01-12 22:02:29 -07:00
|
|
|
val ^= ipv6_addr_hash(&fl6->daddr);
|
|
|
|
val ^= ipv6_addr_hash(&fl6->saddr);
|
2012-10-21 21:42:09 -06:00
|
|
|
|
|
|
|
/* Work only if this not encapsulated */
|
|
|
|
switch (fl6->flowi6_proto) {
|
|
|
|
case IPPROTO_UDP:
|
|
|
|
case IPPROTO_TCP:
|
|
|
|
case IPPROTO_SCTP:
|
2012-10-22 17:35:06 -06:00
|
|
|
val ^= (__force u16)fl6->fl6_sport;
|
|
|
|
val ^= (__force u16)fl6->fl6_dport;
|
2012-10-21 21:42:09 -06:00
|
|
|
break;
|
|
|
|
|
|
|
|
case IPPROTO_ICMPV6:
|
2012-10-22 17:35:06 -06:00
|
|
|
val ^= (__force u16)fl6->fl6_icmp_type;
|
|
|
|
val ^= (__force u16)fl6->fl6_icmp_code;
|
2012-10-21 21:42:09 -06:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
/* RFC6438 recommands to use flowlabel */
|
2012-10-22 17:35:06 -06:00
|
|
|
val ^= (__force u32)fl6->flowlabel;
|
2012-10-21 21:42:09 -06:00
|
|
|
|
|
|
|
/* Perhaps, we need to tune, this function? */
|
|
|
|
val = val ^ (val >> 7) ^ (val >> 12);
|
|
|
|
return val % candidate_count;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct rt6_info *rt6_multipath_select(struct rt6_info *match,
|
2013-06-28 09:35:48 -06:00
|
|
|
struct flowi6 *fl6, int oif,
|
|
|
|
int strict)
|
2012-10-21 21:42:09 -06:00
|
|
|
{
|
|
|
|
struct rt6_info *sibling, *next_sibling;
|
|
|
|
int route_choosen;
|
|
|
|
|
|
|
|
route_choosen = rt6_info_hash_nhsfn(match->rt6i_nsiblings + 1, fl6);
|
|
|
|
/* Don't change the route, if route_choosen == 0
|
|
|
|
* (siblings does not include ourself)
|
|
|
|
*/
|
|
|
|
if (route_choosen)
|
|
|
|
list_for_each_entry_safe(sibling, next_sibling,
|
|
|
|
&match->rt6i_siblings, rt6i_siblings) {
|
|
|
|
route_choosen--;
|
|
|
|
if (route_choosen == 0) {
|
2013-06-28 09:35:48 -06:00
|
|
|
if (rt6_score_route(sibling, oif, strict) < 0)
|
|
|
|
break;
|
2012-10-21 21:42:09 -06:00
|
|
|
match = sibling;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return match;
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
2006-08-05 00:20:06 -06:00
|
|
|
* Route lookup. Any table->tb6_lock is implied.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
static inline struct rt6_info *rt6_device_match(struct net *net,
|
|
|
|
struct rt6_info *rt,
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct in6_addr *saddr,
|
2005-04-16 16:20:36 -06:00
|
|
|
int oif,
|
2008-06-27 21:14:54 -06:00
|
|
|
int flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct rt6_info *local = NULL;
|
|
|
|
struct rt6_info *sprt;
|
|
|
|
|
2008-07-02 03:30:18 -06:00
|
|
|
if (!oif && ipv6_addr_any(saddr))
|
|
|
|
goto out;
|
|
|
|
|
2010-06-11 00:31:35 -06:00
|
|
|
for (sprt = rt; sprt; sprt = sprt->dst.rt6_next) {
|
2011-12-28 18:19:20 -07:00
|
|
|
struct net_device *dev = sprt->dst.dev;
|
2008-07-02 03:30:18 -06:00
|
|
|
|
|
|
|
if (oif) {
|
2005-04-16 16:20:36 -06:00
|
|
|
if (dev->ifindex == oif)
|
|
|
|
return sprt;
|
|
|
|
if (dev->flags & IFF_LOOPBACK) {
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!sprt->rt6i_idev ||
|
2005-04-16 16:20:36 -06:00
|
|
|
sprt->rt6i_idev->dev->ifindex != oif) {
|
2008-06-27 21:14:54 -06:00
|
|
|
if (flags & RT6_LOOKUP_F_IFACE && oif)
|
2005-04-16 16:20:36 -06:00
|
|
|
continue;
|
2007-02-09 07:24:49 -07:00
|
|
|
if (local && (!oif ||
|
2005-04-16 16:20:36 -06:00
|
|
|
local->rt6i_idev->dev->ifindex == oif))
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
local = sprt;
|
|
|
|
}
|
2008-07-02 03:30:18 -06:00
|
|
|
} else {
|
|
|
|
if (ipv6_chk_addr(net, saddr, dev,
|
|
|
|
flags & RT6_LOOKUP_F_IFACE))
|
|
|
|
return sprt;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2008-07-02 03:30:18 -06:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-07-02 03:30:18 -06:00
|
|
|
if (oif) {
|
2005-04-16 16:20:36 -06:00
|
|
|
if (local)
|
|
|
|
return local;
|
|
|
|
|
2008-06-27 21:14:54 -06:00
|
|
|
if (flags & RT6_LOOKUP_F_IFACE)
|
2008-03-04 14:48:30 -07:00
|
|
|
return net->ipv6.ip6_null_entry;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2008-07-02 03:30:18 -06:00
|
|
|
out:
|
2005-04-16 16:20:36 -06:00
|
|
|
return rt;
|
|
|
|
}
|
|
|
|
|
2006-03-20 18:05:13 -07:00
|
|
|
#ifdef CONFIG_IPV6_ROUTER_PREF
|
2013-10-20 22:17:15 -06:00
|
|
|
struct __rt6_probe_work {
|
|
|
|
struct work_struct work;
|
|
|
|
struct in6_addr target;
|
|
|
|
struct net_device *dev;
|
|
|
|
};
|
|
|
|
|
|
|
|
static void rt6_probe_deferred(struct work_struct *w)
|
|
|
|
{
|
|
|
|
struct in6_addr mcaddr;
|
|
|
|
struct __rt6_probe_work *work =
|
|
|
|
container_of(w, struct __rt6_probe_work, work);
|
|
|
|
|
|
|
|
addrconf_addr_solict_mult(&work->target, &mcaddr);
|
|
|
|
ndisc_send_ns(work->dev, NULL, &work->target, &mcaddr, NULL);
|
|
|
|
dev_put(work->dev);
|
|
|
|
kfree(w);
|
|
|
|
}
|
|
|
|
|
2006-03-20 18:05:13 -07:00
|
|
|
static void rt6_probe(struct rt6_info *rt)
|
|
|
|
{
|
2011-07-29 13:00:53 -06:00
|
|
|
struct neighbour *neigh;
|
2006-03-20 18:05:13 -07:00
|
|
|
/*
|
|
|
|
* Okay, this does not seem to be appropriate
|
|
|
|
* for now, however, we need to check if it
|
|
|
|
* is really so; aka Router Reachability Probing.
|
|
|
|
*
|
|
|
|
* Router Reachability Probe MUST be rate-limited
|
|
|
|
* to no more than one per minute.
|
|
|
|
*/
|
2013-01-17 05:53:43 -07:00
|
|
|
if (!rt || !(rt->rt6i_flags & RTF_GATEWAY))
|
2013-01-17 05:53:02 -07:00
|
|
|
return;
|
2013-01-17 05:53:43 -07:00
|
|
|
rcu_read_lock_bh();
|
|
|
|
neigh = __ipv6_neigh_lookup_noref(rt->dst.dev, &rt->rt6i_gateway);
|
|
|
|
if (neigh) {
|
|
|
|
write_lock(&neigh->lock);
|
|
|
|
if (neigh->nud_state & NUD_VALID)
|
|
|
|
goto out;
|
2013-01-17 05:53:02 -07:00
|
|
|
}
|
2013-01-17 05:53:43 -07:00
|
|
|
|
|
|
|
if (!neigh ||
|
2006-03-20 18:05:47 -07:00
|
|
|
time_after(jiffies, neigh->updated + rt->rt6i_idev->cnf.rtr_probe_interval)) {
|
2013-10-20 22:17:15 -06:00
|
|
|
struct __rt6_probe_work *work;
|
2006-03-20 18:05:13 -07:00
|
|
|
|
2013-10-20 22:17:15 -06:00
|
|
|
work = kmalloc(sizeof(*work), GFP_ATOMIC);
|
|
|
|
|
|
|
|
if (neigh && work)
|
2013-01-21 02:58:50 -07:00
|
|
|
neigh->updated = jiffies;
|
2013-10-20 22:17:15 -06:00
|
|
|
|
|
|
|
if (neigh)
|
2013-01-17 05:53:43 -07:00
|
|
|
write_unlock(&neigh->lock);
|
|
|
|
|
2013-10-20 22:17:15 -06:00
|
|
|
if (work) {
|
|
|
|
INIT_WORK(&work->work, rt6_probe_deferred);
|
|
|
|
work->target = rt->rt6i_gateway;
|
|
|
|
dev_hold(rt->dst.dev);
|
|
|
|
work->dev = rt->dst.dev;
|
|
|
|
schedule_work(&work->work);
|
|
|
|
}
|
2011-07-29 13:00:53 -06:00
|
|
|
} else {
|
2013-01-17 05:53:43 -07:00
|
|
|
out:
|
|
|
|
write_unlock(&neigh->lock);
|
2011-07-29 13:00:53 -06:00
|
|
|
}
|
2013-01-17 05:53:43 -07:00
|
|
|
rcu_read_unlock_bh();
|
2006-03-20 18:05:13 -07:00
|
|
|
}
|
|
|
|
#else
|
|
|
|
static inline void rt6_probe(struct rt6_info *rt)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
2006-03-20 18:00:26 -07:00
|
|
|
* Default Router Selection (RFC 2461 6.3.6)
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2007-03-22 13:27:49 -06:00
|
|
|
static inline int rt6_check_dev(struct rt6_info *rt, int oif)
|
2006-03-20 18:00:26 -07:00
|
|
|
{
|
2011-12-28 18:19:20 -07:00
|
|
|
struct net_device *dev = rt->dst.dev;
|
2007-04-06 12:42:27 -06:00
|
|
|
if (!oif || dev->ifindex == oif)
|
2006-03-20 18:00:26 -07:00
|
|
|
return 2;
|
2007-04-06 12:42:27 -06:00
|
|
|
if ((dev->flags & IFF_LOOPBACK) &&
|
|
|
|
rt->rt6i_idev && rt->rt6i_idev->dev->ifindex == oif)
|
|
|
|
return 1;
|
|
|
|
return 0;
|
2006-03-20 18:00:26 -07:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2013-07-11 04:43:42 -06:00
|
|
|
static inline enum rt6_nud_state rt6_check_neigh(struct rt6_info *rt)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2011-07-29 13:00:53 -06:00
|
|
|
struct neighbour *neigh;
|
2013-07-11 04:43:42 -06:00
|
|
|
enum rt6_nud_state ret = RT6_NUD_FAIL_HARD;
|
2011-07-29 13:00:53 -06:00
|
|
|
|
2006-05-26 14:23:41 -06:00
|
|
|
if (rt->rt6i_flags & RTF_NONEXTHOP ||
|
|
|
|
!(rt->rt6i_flags & RTF_GATEWAY))
|
2013-07-11 04:43:42 -06:00
|
|
|
return RT6_NUD_SUCCEED;
|
2013-01-17 05:53:38 -07:00
|
|
|
|
|
|
|
rcu_read_lock_bh();
|
|
|
|
neigh = __ipv6_neigh_lookup_noref(rt->dst.dev, &rt->rt6i_gateway);
|
|
|
|
if (neigh) {
|
|
|
|
read_lock(&neigh->lock);
|
2006-03-20 18:00:26 -07:00
|
|
|
if (neigh->nud_state & NUD_VALID)
|
2013-07-11 04:43:42 -06:00
|
|
|
ret = RT6_NUD_SUCCEED;
|
2008-01-19 01:35:16 -07:00
|
|
|
#ifdef CONFIG_IPV6_ROUTER_PREF
|
2012-12-03 03:26:54 -07:00
|
|
|
else if (!(neigh->nud_state & NUD_FAILED))
|
2013-07-11 04:43:42 -06:00
|
|
|
ret = RT6_NUD_SUCCEED;
|
2008-01-19 01:35:16 -07:00
|
|
|
#endif
|
2013-01-17 05:53:38 -07:00
|
|
|
read_unlock(&neigh->lock);
|
2013-07-11 04:43:42 -06:00
|
|
|
} else {
|
|
|
|
ret = IS_ENABLED(CONFIG_IPV6_ROUTER_PREF) ?
|
|
|
|
RT6_NUD_SUCCEED : RT6_NUD_FAIL_SOFT;
|
2012-12-03 03:26:54 -07:00
|
|
|
}
|
2013-01-17 05:53:38 -07:00
|
|
|
rcu_read_unlock_bh();
|
|
|
|
|
2012-12-03 03:26:54 -07:00
|
|
|
return ret;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-03-20 18:00:26 -07:00
|
|
|
static int rt6_score_route(struct rt6_info *rt, int oif,
|
|
|
|
int strict)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2012-12-03 03:26:54 -07:00
|
|
|
int m;
|
2007-02-09 07:24:49 -07:00
|
|
|
|
2006-05-26 14:23:41 -06:00
|
|
|
m = rt6_check_dev(rt, oif);
|
2006-08-23 18:25:05 -06:00
|
|
|
if (!m && (strict & RT6_LOOKUP_F_IFACE))
|
2013-07-11 04:43:42 -06:00
|
|
|
return RT6_NUD_FAIL_HARD;
|
2006-03-20 18:04:53 -07:00
|
|
|
#ifdef CONFIG_IPV6_ROUTER_PREF
|
|
|
|
m |= IPV6_DECODE_PREF(IPV6_EXTRACT_PREF(rt->rt6i_flags)) << 2;
|
|
|
|
#endif
|
2013-07-11 04:43:42 -06:00
|
|
|
if (strict & RT6_LOOKUP_F_REACHABLE) {
|
|
|
|
int n = rt6_check_neigh(rt);
|
|
|
|
if (n < 0)
|
|
|
|
return n;
|
|
|
|
}
|
2006-03-20 18:00:26 -07:00
|
|
|
return m;
|
|
|
|
}
|
|
|
|
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
static struct rt6_info *find_match(struct rt6_info *rt, int oif, int strict,
|
2013-07-11 04:43:42 -06:00
|
|
|
int *mpri, struct rt6_info *match,
|
|
|
|
bool *do_rr)
|
2006-03-20 18:00:26 -07:00
|
|
|
{
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
int m;
|
2013-07-11 04:43:42 -06:00
|
|
|
bool match_do_rr = false;
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
|
|
|
|
if (rt6_check_expired(rt))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
m = rt6_score_route(rt, oif, strict);
|
2013-10-30 01:39:26 -06:00
|
|
|
if (m == RT6_NUD_FAIL_SOFT) {
|
2013-07-11 04:43:42 -06:00
|
|
|
match_do_rr = true;
|
|
|
|
m = 0; /* lowest valid score */
|
|
|
|
} else if (m < 0) {
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
goto out;
|
2013-07-11 04:43:42 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
if (strict & RT6_LOOKUP_F_REACHABLE)
|
|
|
|
rt6_probe(rt);
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
|
|
|
|
if (m > *mpri) {
|
2013-07-11 04:43:42 -06:00
|
|
|
*do_rr = match_do_rr;
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
*mpri = m;
|
|
|
|
match = rt;
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
return match;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct rt6_info *find_rr_leaf(struct fib6_node *fn,
|
|
|
|
struct rt6_info *rr_head,
|
2013-07-11 04:43:42 -06:00
|
|
|
u32 metric, int oif, int strict,
|
|
|
|
bool *do_rr)
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
{
|
|
|
|
struct rt6_info *rt, *match;
|
2006-03-20 18:00:26 -07:00
|
|
|
int mpri = -1;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
match = NULL;
|
|
|
|
for (rt = rr_head; rt && rt->rt6i_metric == metric;
|
2010-06-11 00:31:35 -06:00
|
|
|
rt = rt->dst.rt6_next)
|
2013-07-11 04:43:42 -06:00
|
|
|
match = find_match(rt, oif, strict, &mpri, match, do_rr);
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
for (rt = fn->leaf; rt && rt != rr_head && rt->rt6i_metric == metric;
|
2010-06-11 00:31:35 -06:00
|
|
|
rt = rt->dst.rt6_next)
|
2013-07-11 04:43:42 -06:00
|
|
|
match = find_match(rt, oif, strict, &mpri, match, do_rr);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
return match;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
|
|
|
|
{
|
|
|
|
struct rt6_info *match, *rt0;
|
2008-03-04 14:48:30 -07:00
|
|
|
struct net *net;
|
2013-07-11 04:43:42 -06:00
|
|
|
bool do_rr = false;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
rt0 = fn->rr_ptr;
|
|
|
|
if (!rt0)
|
|
|
|
fn->rr_ptr = rt0 = fn->leaf;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2013-07-11 04:43:42 -06:00
|
|
|
match = find_rr_leaf(fn, rt0, rt0->rt6i_metric, oif, strict,
|
|
|
|
&do_rr);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2013-07-11 04:43:42 -06:00
|
|
|
if (do_rr) {
|
2010-06-11 00:31:35 -06:00
|
|
|
struct rt6_info *next = rt0->dst.rt6_next;
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
|
2006-03-20 18:00:26 -07:00
|
|
|
/* no entries matched; do round-robin */
|
[IPV6]: Fix routing round-robin locking.
As per RFC2461, section 6.3.6, item #2, when no routers on the
matching list are known to be reachable or probably reachable we
do round robin on those available routes so that we make sure
to probe as many of them as possible to detect when one becomes
reachable faster.
Each routing table has a rwlock protecting the tree and the linked
list of routes at each leaf. The round robin code executes during
lookup and thus with the rwlock taken as a reader. A small local
spinlock tries to provide protection but this does not work at all
for two reasons:
1) The round-robin list manipulation, as coded, goes like this (with
read lock held):
walk routes finding head and tail
spin_lock();
rotate list using head and tail
spin_unlock();
While one thread is rotating the list, another thread can
end up with stale values of head and tail and then proceed
to corrupt the list when it gets the lock. This ends up causing
the OOPS in fib6_add() later onthat many people have been hitting.
2) All the other code paths that run with the rwlock held as
a reader do not expect the list to change on them, they
expect it to remain completely fixed while they hold the
lock in that way.
So, simply stated, it is impossible to implement this correctly using
a manipulation of the list without violating the rwlock locking
semantics.
Reimplement using a per-fib6_node round-robin pointer. This way we
don't need to manipulate the list at all, and since the round-robin
pointer can only ever point to real existing entries we don't need
to perform any locking on the changing of the round-robin pointer
itself. We only need to reset the round-robin pointer to NULL when
the entry it is pointing to is removed.
The idea is from Thomas Graf and it is very similar to how this
was implemented before the advanced router selection code when in.
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-24 21:36:25 -06:00
|
|
|
if (!next || next->rt6i_metric != rt0->rt6i_metric)
|
|
|
|
next = fn->leaf;
|
|
|
|
|
|
|
|
if (next != rt0)
|
|
|
|
fn->rr_ptr = next;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2011-12-28 18:19:20 -07:00
|
|
|
net = dev_net(rt0->dst.dev);
|
2010-09-22 14:43:57 -06:00
|
|
|
return match ? match : net->ipv6.ip6_null_entry;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-03-20 18:06:24 -07:00
|
|
|
#ifdef CONFIG_IPV6_ROUTE_INFO
|
|
|
|
int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct in6_addr *gwaddr)
|
2006-03-20 18:06:24 -07:00
|
|
|
{
|
2008-03-25 06:47:49 -06:00
|
|
|
struct net *net = dev_net(dev);
|
2006-03-20 18:06:24 -07:00
|
|
|
struct route_info *rinfo = (struct route_info *) opt;
|
|
|
|
struct in6_addr prefix_buf, *prefix;
|
|
|
|
unsigned int pref;
|
2008-05-27 02:37:49 -06:00
|
|
|
unsigned long lifetime;
|
2006-03-20 18:06:24 -07:00
|
|
|
struct rt6_info *rt;
|
|
|
|
|
|
|
|
if (len < sizeof(struct route_info)) {
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Sanity check for prefix_len and length */
|
|
|
|
if (rinfo->length > 3) {
|
|
|
|
return -EINVAL;
|
|
|
|
} else if (rinfo->prefix_len > 128) {
|
|
|
|
return -EINVAL;
|
|
|
|
} else if (rinfo->prefix_len > 64) {
|
|
|
|
if (rinfo->length < 2) {
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
} else if (rinfo->prefix_len > 0) {
|
|
|
|
if (rinfo->length < 1) {
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
pref = rinfo->route_pref;
|
|
|
|
if (pref == ICMPV6_ROUTER_PREF_INVALID)
|
2009-09-10 00:25:11 -06:00
|
|
|
return -EINVAL;
|
2006-03-20 18:06:24 -07:00
|
|
|
|
2008-05-27 02:37:49 -06:00
|
|
|
lifetime = addrconf_timeout_fixup(ntohl(rinfo->lifetime), HZ);
|
2006-03-20 18:06:24 -07:00
|
|
|
|
|
|
|
if (rinfo->length == 3)
|
|
|
|
prefix = (struct in6_addr *)rinfo->prefix;
|
|
|
|
else {
|
|
|
|
/* this function is safe */
|
|
|
|
ipv6_addr_prefix(&prefix_buf,
|
|
|
|
(struct in6_addr *)rinfo->prefix,
|
|
|
|
rinfo->prefix_len);
|
|
|
|
prefix = &prefix_buf;
|
|
|
|
}
|
|
|
|
|
2013-11-07 18:56:53 -07:00
|
|
|
if (rinfo->prefix_len == 0)
|
|
|
|
rt = rt6_get_dflt_router(gwaddr, dev);
|
|
|
|
else
|
|
|
|
rt = rt6_get_route_info(net, prefix, rinfo->prefix_len,
|
|
|
|
gwaddr, dev->ifindex);
|
2006-03-20 18:06:24 -07:00
|
|
|
|
|
|
|
if (rt && !lifetime) {
|
2006-08-22 01:00:21 -06:00
|
|
|
ip6_del_rt(rt);
|
2006-03-20 18:06:24 -07:00
|
|
|
rt = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!rt && lifetime)
|
2008-03-04 14:46:48 -07:00
|
|
|
rt = rt6_add_route_info(net, prefix, rinfo->prefix_len, gwaddr, dev->ifindex,
|
2006-03-20 18:06:24 -07:00
|
|
|
pref);
|
|
|
|
else if (rt)
|
|
|
|
rt->rt6i_flags = RTF_ROUTEINFO |
|
|
|
|
(rt->rt6i_flags & ~RTF_PREF_MASK) | RTF_PREF(pref);
|
|
|
|
|
|
|
|
if (rt) {
|
2012-04-05 18:13:10 -06:00
|
|
|
if (!addrconf_finite_timeout(lifetime))
|
|
|
|
rt6_clean_expires(rt);
|
|
|
|
else
|
|
|
|
rt6_set_expires(rt, jiffies + HZ * lifetime);
|
|
|
|
|
2012-10-28 18:13:19 -06:00
|
|
|
ip6_rt_put(rt);
|
2006-03-20 18:06:24 -07:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
#define BACKTRACK(__net, saddr) \
|
2006-08-23 18:22:39 -06:00
|
|
|
do { \
|
2008-03-04 14:48:30 -07:00
|
|
|
if (rt == __net->ipv6.ip6_null_entry) { \
|
2006-08-23 18:22:39 -06:00
|
|
|
struct fib6_node *pn; \
|
2006-10-16 23:11:11 -06:00
|
|
|
while (1) { \
|
2006-08-23 18:22:39 -06:00
|
|
|
if (fn->fn_flags & RTN_TL_ROOT) \
|
|
|
|
goto out; \
|
|
|
|
pn = fn->parent; \
|
|
|
|
if (FIB6_SUBTREE(pn) && FIB6_SUBTREE(pn) != fn) \
|
2006-12-13 17:38:29 -07:00
|
|
|
fn = fib6_lookup(FIB6_SUBTREE(pn), NULL, saddr); \
|
2006-08-23 18:22:39 -06:00
|
|
|
else \
|
|
|
|
fn = pn; \
|
|
|
|
if (fn->fn_flags & RTN_RTINFO) \
|
|
|
|
goto restart; \
|
2006-08-05 00:20:06 -06:00
|
|
|
} \
|
|
|
|
} \
|
2011-12-03 16:02:47 -07:00
|
|
|
} while (0)
|
2006-08-05 00:20:06 -06:00
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
static struct rt6_info *ip6_pol_route_lookup(struct net *net,
|
|
|
|
struct fib6_table *table,
|
2011-03-12 14:22:43 -07:00
|
|
|
struct flowi6 *fl6, int flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct fib6_node *fn;
|
|
|
|
struct rt6_info *rt;
|
|
|
|
|
2006-08-05 00:20:06 -06:00
|
|
|
read_lock_bh(&table->tb6_lock);
|
2011-03-12 14:22:43 -07:00
|
|
|
fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr);
|
2006-08-05 00:20:06 -06:00
|
|
|
restart:
|
|
|
|
rt = fn->leaf;
|
2011-03-12 14:22:43 -07:00
|
|
|
rt = rt6_device_match(net, rt, &fl6->saddr, fl6->flowi6_oif, flags);
|
2012-10-21 21:42:09 -06:00
|
|
|
if (rt->rt6i_nsiblings && fl6->flowi6_oif == 0)
|
2013-06-28 09:35:48 -06:00
|
|
|
rt = rt6_multipath_select(rt, fl6, fl6->flowi6_oif, flags);
|
2011-03-12 14:22:43 -07:00
|
|
|
BACKTRACK(net, &fl6->saddr);
|
2006-08-05 00:20:06 -06:00
|
|
|
out:
|
2010-06-11 00:31:35 -06:00
|
|
|
dst_use(&rt->dst, jiffies);
|
2006-08-05 00:20:06 -06:00
|
|
|
read_unlock_bh(&table->tb6_lock);
|
|
|
|
return rt;
|
|
|
|
|
|
|
|
}
|
|
|
|
|
2011-09-05 08:05:44 -06:00
|
|
|
struct dst_entry * ip6_route_lookup(struct net *net, struct flowi6 *fl6,
|
|
|
|
int flags)
|
|
|
|
{
|
|
|
|
return fib6_rule_lookup(net, fl6, flags, ip6_pol_route_lookup);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(ip6_route_lookup);
|
|
|
|
|
[IPV6]: Make address arguments const.
- net/ipv6/addrconf.c:
ipv6_get_ifaddr(), ipv6_dev_get_saddr()
- net/ipv6/mcast.c:
ipv6_sock_mc_join(), ipv6_sock_mc_drop(),
inet6_mc_check(),
ipv6_dev_mc_inc(), __ipv6_dev_mc_dec(), ipv6_dev_mc_dec(),
ipv6_chk_mcast_addr()
- net/ipv6/route.c:
rt6_lookup(), icmp6_dst_alloc()
- net/ipv6/ip6_output.c:
ip6_nd_hdr()
- net/ipv6/ndisc.c:
ndisc_send_ns(), ndisc_send_rs(), ndisc_send_redirect(),
ndisc_get_neigh(), __ndisc_send()
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
2008-04-10 00:42:10 -06:00
|
|
|
struct rt6_info *rt6_lookup(struct net *net, const struct in6_addr *daddr,
|
|
|
|
const struct in6_addr *saddr, int oif, int strict)
|
2006-08-05 00:20:06 -06:00
|
|
|
{
|
2011-03-12 14:22:43 -07:00
|
|
|
struct flowi6 fl6 = {
|
|
|
|
.flowi6_oif = oif,
|
|
|
|
.daddr = *daddr,
|
2006-08-05 00:20:06 -06:00
|
|
|
};
|
|
|
|
struct dst_entry *dst;
|
2006-08-23 18:25:05 -06:00
|
|
|
int flags = strict ? RT6_LOOKUP_F_IFACE : 0;
|
2006-08-05 00:20:06 -06:00
|
|
|
|
2006-10-13 16:01:03 -06:00
|
|
|
if (saddr) {
|
2011-03-12 14:22:43 -07:00
|
|
|
memcpy(&fl6.saddr, saddr, sizeof(*saddr));
|
2006-10-13 16:01:03 -06:00
|
|
|
flags |= RT6_LOOKUP_F_HAS_SADDR;
|
|
|
|
}
|
|
|
|
|
2011-03-12 14:22:43 -07:00
|
|
|
dst = fib6_rule_lookup(net, &fl6, flags, ip6_pol_route_lookup);
|
2006-08-05 00:20:06 -06:00
|
|
|
if (dst->error == 0)
|
|
|
|
return (struct rt6_info *) dst;
|
|
|
|
|
|
|
|
dst_release(dst);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2007-02-22 06:05:40 -07:00
|
|
|
EXPORT_SYMBOL(rt6_lookup);
|
|
|
|
|
2006-08-05 00:20:06 -06:00
|
|
|
/* ip6_ins_rt is called with FREE table->tb6_lock.
|
2005-04-16 16:20:36 -06:00
|
|
|
It takes new route entry, the addition fails by any reason the
|
|
|
|
route is freed. In any case, if caller does not hold it, it may
|
|
|
|
be destroyed.
|
|
|
|
*/
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
int err;
|
2006-08-05 00:20:06 -06:00
|
|
|
struct fib6_table *table;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-05 00:20:06 -06:00
|
|
|
table = rt->rt6i_table;
|
|
|
|
write_lock_bh(&table->tb6_lock);
|
2006-08-22 01:01:08 -06:00
|
|
|
err = fib6_add(&table->tb6_root, rt, info);
|
2006-08-05 00:20:06 -06:00
|
|
|
write_unlock_bh(&table->tb6_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2006-08-22 01:00:45 -06:00
|
|
|
int ip6_ins_rt(struct rt6_info *rt)
|
|
|
|
{
|
2008-01-10 04:26:13 -07:00
|
|
|
struct nl_info info = {
|
2011-12-28 18:19:20 -07:00
|
|
|
.nl_net = dev_net(rt->dst.dev),
|
2008-01-10 04:26:13 -07:00
|
|
|
};
|
2007-12-13 10:45:12 -07:00
|
|
|
return __ip6_ins_rt(rt, &info);
|
2006-08-22 01:00:45 -06:00
|
|
|
}
|
|
|
|
|
2012-04-05 18:13:10 -06:00
|
|
|
static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort,
|
2011-07-19 14:18:36 -06:00
|
|
|
const struct in6_addr *daddr,
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct in6_addr *saddr)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct rt6_info *rt;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Clone the route.
|
|
|
|
*/
|
|
|
|
|
2011-07-19 14:18:36 -06:00
|
|
|
rt = ip6_rt_copy(ort, daddr);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
if (rt) {
|
2013-11-04 22:34:53 -07:00
|
|
|
if (ort->rt6i_dst.plen != 128 &&
|
|
|
|
ipv6_addr_equal(&ort->rt6i_dst.addr, daddr))
|
|
|
|
rt->rt6i_flags |= RTF_ANYCAST;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
rt->rt6i_flags |= RTF_CACHE;
|
|
|
|
|
|
|
|
#ifdef CONFIG_IPV6_SUBTREES
|
|
|
|
if (rt->rt6i_src.plen && saddr) {
|
2011-11-20 20:39:03 -07:00
|
|
|
rt->rt6i_src.addr = *saddr;
|
2005-04-16 16:20:36 -06:00
|
|
|
rt->rt6i_src.plen = 128;
|
|
|
|
}
|
|
|
|
#endif
|
2006-03-20 17:55:51 -07:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-03-20 17:55:51 -07:00
|
|
|
return rt;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-07-19 14:18:36 -06:00
|
|
|
static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort,
|
|
|
|
const struct in6_addr *daddr)
|
2006-03-20 17:58:32 -07:00
|
|
|
{
|
2011-07-19 14:18:36 -06:00
|
|
|
struct rt6_info *rt = ip6_rt_copy(ort, daddr);
|
|
|
|
|
2013-01-17 05:54:05 -07:00
|
|
|
if (rt)
|
2006-03-20 17:58:32 -07:00
|
|
|
rt->rt6i_flags |= RTF_CACHE;
|
|
|
|
return rt;
|
|
|
|
}
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
static struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, int oif,
|
2011-03-12 14:22:43 -07:00
|
|
|
struct flowi6 *fl6, int flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct fib6_node *fn;
|
2006-03-20 18:00:05 -07:00
|
|
|
struct rt6_info *rt, *nrt;
|
2006-08-05 00:20:06 -06:00
|
|
|
int strict = 0;
|
2005-04-16 16:20:36 -06:00
|
|
|
int attempts = 3;
|
2006-03-20 18:00:05 -07:00
|
|
|
int err;
|
2008-07-19 23:35:03 -06:00
|
|
|
int reachable = net->ipv6.devconf_all->forwarding ? 0 : RT6_LOOKUP_F_REACHABLE;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-23 18:25:05 -06:00
|
|
|
strict |= flags & RT6_LOOKUP_F_IFACE;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
relookup:
|
2006-08-05 00:20:06 -06:00
|
|
|
read_lock_bh(&table->tb6_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-03-20 18:04:35 -07:00
|
|
|
restart_2:
|
2011-03-12 14:22:43 -07:00
|
|
|
fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
restart:
|
2007-10-15 14:02:51 -06:00
|
|
|
rt = rt6_select(fn, oif, strict | reachable);
|
2013-06-28 09:35:48 -06:00
|
|
|
if (rt->rt6i_nsiblings)
|
|
|
|
rt = rt6_multipath_select(rt, fl6, oif, strict | reachable);
|
2011-03-12 14:22:43 -07:00
|
|
|
BACKTRACK(net, &fl6->saddr);
|
2008-03-04 14:48:30 -07:00
|
|
|
if (rt == net->ipv6.ip6_null_entry ||
|
2006-03-20 18:04:35 -07:00
|
|
|
rt->rt6i_flags & RTF_CACHE)
|
2006-03-20 18:01:24 -07:00
|
|
|
goto out;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-06-11 00:31:35 -06:00
|
|
|
dst_hold(&rt->dst);
|
2006-08-05 00:20:06 -06:00
|
|
|
read_unlock_bh(&table->tb6_lock);
|
2006-03-20 17:59:08 -07:00
|
|
|
|
2013-01-17 05:53:32 -07:00
|
|
|
if (!(rt->rt6i_flags & (RTF_NONEXTHOP | RTF_GATEWAY)))
|
2011-03-12 14:22:43 -07:00
|
|
|
nrt = rt6_alloc_cow(rt, &fl6->daddr, &fl6->saddr);
|
2011-03-09 20:55:25 -07:00
|
|
|
else if (!(rt->dst.flags & DST_HOST))
|
2011-03-12 14:22:43 -07:00
|
|
|
nrt = rt6_alloc_clone(rt, &fl6->daddr);
|
2011-03-09 20:55:25 -07:00
|
|
|
else
|
|
|
|
goto out2;
|
2006-03-20 17:59:27 -07:00
|
|
|
|
2012-10-28 18:13:19 -06:00
|
|
|
ip6_rt_put(rt);
|
2008-03-04 14:48:30 -07:00
|
|
|
rt = nrt ? : net->ipv6.ip6_null_entry;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-06-11 00:31:35 -06:00
|
|
|
dst_hold(&rt->dst);
|
2006-03-20 18:00:05 -07:00
|
|
|
if (nrt) {
|
2006-08-22 01:00:45 -06:00
|
|
|
err = ip6_ins_rt(nrt);
|
2006-03-20 18:00:05 -07:00
|
|
|
if (!err)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out2;
|
|
|
|
}
|
|
|
|
|
2006-03-20 18:00:05 -07:00
|
|
|
if (--attempts <= 0)
|
|
|
|
goto out2;
|
|
|
|
|
|
|
|
/*
|
2006-08-05 00:20:06 -06:00
|
|
|
* Race condition! In the gap, when table->tb6_lock was
|
2006-03-20 18:00:05 -07:00
|
|
|
* released someone could insert this route. Relookup.
|
|
|
|
*/
|
2012-10-28 18:13:19 -06:00
|
|
|
ip6_rt_put(rt);
|
2006-03-20 18:00:05 -07:00
|
|
|
goto relookup;
|
|
|
|
|
|
|
|
out:
|
2006-03-20 18:04:35 -07:00
|
|
|
if (reachable) {
|
|
|
|
reachable = 0;
|
|
|
|
goto restart_2;
|
|
|
|
}
|
2010-06-11 00:31:35 -06:00
|
|
|
dst_hold(&rt->dst);
|
2006-08-05 00:20:06 -06:00
|
|
|
read_unlock_bh(&table->tb6_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
out2:
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.lastuse = jiffies;
|
|
|
|
rt->dst.__use++;
|
2006-08-05 00:20:06 -06:00
|
|
|
|
|
|
|
return rt;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
static struct rt6_info *ip6_pol_route_input(struct net *net, struct fib6_table *table,
|
2011-03-12 14:22:43 -07:00
|
|
|
struct flowi6 *fl6, int flags)
|
2007-10-15 14:02:51 -06:00
|
|
|
{
|
2011-03-12 14:22:43 -07:00
|
|
|
return ip6_pol_route(net, table, fl6->flowi6_iif, fl6, flags);
|
2007-10-15 14:02:51 -06:00
|
|
|
}
|
|
|
|
|
2012-03-31 22:03:45 -06:00
|
|
|
static struct dst_entry *ip6_route_input_lookup(struct net *net,
|
|
|
|
struct net_device *dev,
|
|
|
|
struct flowi6 *fl6, int flags)
|
|
|
|
{
|
|
|
|
if (rt6_need_strict(&fl6->daddr) && dev->type != ARPHRD_PIMREG)
|
|
|
|
flags |= RT6_LOOKUP_F_IFACE;
|
|
|
|
|
|
|
|
return fib6_rule_lookup(net, fl6, flags, ip6_pol_route_input);
|
|
|
|
}
|
|
|
|
|
2006-08-05 00:20:06 -06:00
|
|
|
void ip6_route_input(struct sk_buff *skb)
|
|
|
|
{
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct ipv6hdr *iph = ipv6_hdr(skb);
|
2008-03-25 06:47:49 -06:00
|
|
|
struct net *net = dev_net(skb->dev);
|
2006-10-13 16:01:03 -06:00
|
|
|
int flags = RT6_LOOKUP_F_HAS_SADDR;
|
2011-03-12 14:22:43 -07:00
|
|
|
struct flowi6 fl6 = {
|
|
|
|
.flowi6_iif = skb->dev->ifindex,
|
|
|
|
.daddr = iph->daddr,
|
|
|
|
.saddr = iph->saddr,
|
2013-01-12 22:01:51 -07:00
|
|
|
.flowlabel = ip6_flowinfo(iph),
|
2011-03-12 14:22:43 -07:00
|
|
|
.flowi6_mark = skb->mark,
|
|
|
|
.flowi6_proto = iph->nexthdr,
|
2006-08-05 00:20:06 -06:00
|
|
|
};
|
2006-10-13 16:01:03 -06:00
|
|
|
|
2012-03-31 22:03:45 -06:00
|
|
|
skb_dst_set(skb, ip6_route_input_lookup(net, skb->dev, &fl6, flags));
|
2006-08-05 00:20:06 -06:00
|
|
|
}
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
static struct rt6_info *ip6_pol_route_output(struct net *net, struct fib6_table *table,
|
2011-03-12 14:22:43 -07:00
|
|
|
struct flowi6 *fl6, int flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2011-03-12 14:22:43 -07:00
|
|
|
return ip6_pol_route(net, table, fl6->flowi6_oif, fl6, flags);
|
2006-08-05 00:20:06 -06:00
|
|
|
}
|
|
|
|
|
2011-03-22 20:17:36 -06:00
|
|
|
struct dst_entry * ip6_route_output(struct net *net, const struct sock *sk,
|
2011-03-12 14:22:43 -07:00
|
|
|
struct flowi6 *fl6)
|
2006-08-05 00:20:06 -06:00
|
|
|
{
|
|
|
|
int flags = 0;
|
|
|
|
|
2012-08-08 15:53:36 -06:00
|
|
|
fl6->flowi6_iif = LOOPBACK_IFINDEX;
|
2012-06-25 09:42:26 -06:00
|
|
|
|
2011-03-12 14:22:43 -07:00
|
|
|
if ((sk && sk->sk_bound_dev_if) || rt6_need_strict(&fl6->daddr))
|
2006-08-23 18:25:05 -06:00
|
|
|
flags |= RT6_LOOKUP_F_IFACE;
|
2006-08-05 00:20:06 -06:00
|
|
|
|
2011-03-12 14:22:43 -07:00
|
|
|
if (!ipv6_addr_any(&fl6->saddr))
|
2006-10-13 16:01:03 -06:00
|
|
|
flags |= RT6_LOOKUP_F_HAS_SADDR;
|
2010-03-06 17:14:44 -07:00
|
|
|
else if (sk)
|
|
|
|
flags |= rt6_srcprefs2flags(inet6_sk(sk)->srcprefs);
|
2006-10-13 16:01:03 -06:00
|
|
|
|
2011-03-12 14:22:43 -07:00
|
|
|
return fib6_rule_lookup(net, fl6, flags, ip6_pol_route_output);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2007-02-22 06:05:40 -07:00
|
|
|
EXPORT_SYMBOL(ip6_route_output);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-03-01 15:59:04 -07:00
|
|
|
struct dst_entry *ip6_blackhole_route(struct net *net, struct dst_entry *dst_orig)
|
2007-05-24 19:17:54 -06:00
|
|
|
{
|
2011-04-28 15:13:38 -06:00
|
|
|
struct rt6_info *rt, *ort = (struct rt6_info *) dst_orig;
|
2007-05-24 19:17:54 -06:00
|
|
|
struct dst_entry *new = NULL;
|
|
|
|
|
2012-07-19 13:31:33 -06:00
|
|
|
rt = dst_alloc(&ip6_dst_blackhole_ops, ort->dst.dev, 1, DST_OBSOLETE_NONE, 0);
|
2007-05-24 19:17:54 -06:00
|
|
|
if (rt) {
|
2010-06-11 00:31:35 -06:00
|
|
|
new = &rt->dst;
|
2007-05-24 19:17:54 -06:00
|
|
|
|
2012-07-05 17:37:09 -06:00
|
|
|
memset(new + 1, 0, sizeof(*rt) - sizeof(*new));
|
|
|
|
rt6_init_peer(rt, net->ipv6.peers);
|
|
|
|
|
2007-05-24 19:17:54 -06:00
|
|
|
new->__use = 1;
|
2007-11-13 22:34:06 -07:00
|
|
|
new->input = dst_discard;
|
|
|
|
new->output = dst_discard;
|
2007-05-24 19:17:54 -06:00
|
|
|
|
2011-07-19 14:18:36 -06:00
|
|
|
if (dst_metrics_read_only(&ort->dst))
|
|
|
|
new->_metrics = ort->dst._metrics;
|
|
|
|
else
|
|
|
|
dst_copy_metrics(new, &ort->dst);
|
2007-05-24 19:17:54 -06:00
|
|
|
rt->rt6i_idev = ort->rt6i_idev;
|
|
|
|
if (rt->rt6i_idev)
|
|
|
|
in6_dev_hold(rt->rt6i_idev);
|
|
|
|
|
2011-11-20 20:39:03 -07:00
|
|
|
rt->rt6i_gateway = ort->rt6i_gateway;
|
2012-04-05 18:13:10 -06:00
|
|
|
rt->rt6i_flags = ort->rt6i_flags;
|
2007-05-24 19:17:54 -06:00
|
|
|
rt->rt6i_metric = 0;
|
|
|
|
|
|
|
|
memcpy(&rt->rt6i_dst, &ort->rt6i_dst, sizeof(struct rt6key));
|
|
|
|
#ifdef CONFIG_IPV6_SUBTREES
|
|
|
|
memcpy(&rt->rt6i_src, &ort->rt6i_src, sizeof(struct rt6key));
|
|
|
|
#endif
|
|
|
|
|
|
|
|
dst_free(new);
|
|
|
|
}
|
|
|
|
|
2011-03-01 15:45:33 -07:00
|
|
|
dst_release(dst_orig);
|
|
|
|
return new ? new : ERR_PTR(-ENOMEM);
|
2007-05-24 19:17:54 -06:00
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Destination cache support functions
|
|
|
|
*/
|
|
|
|
|
|
|
|
static struct dst_entry *ip6_dst_check(struct dst_entry *dst, u32 cookie)
|
|
|
|
{
|
|
|
|
struct rt6_info *rt;
|
|
|
|
|
|
|
|
rt = (struct rt6_info *) dst;
|
|
|
|
|
2012-09-10 16:09:46 -06:00
|
|
|
/* All IPV6 dsts are created with ->obsolete set to the value
|
|
|
|
* DST_OBSOLETE_FORCE_CHK which forces validation calls down
|
|
|
|
* into this function always.
|
|
|
|
*/
|
2013-07-29 18:33:53 -06:00
|
|
|
if (rt->rt6i_genid != rt_genid_ipv6(dev_net(rt->dst.dev)))
|
2012-09-10 16:09:46 -06:00
|
|
|
return NULL;
|
|
|
|
|
2013-10-23 23:48:24 -06:00
|
|
|
if (!rt->rt6i_node || (rt->rt6i_node->fn_sernum != cookie))
|
|
|
|
return NULL;
|
2012-11-07 14:56:33 -07:00
|
|
|
|
2013-10-23 23:48:24 -06:00
|
|
|
if (rt6_check_expired(rt))
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
return dst;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct dst_entry *ip6_negative_advice(struct dst_entry *dst)
|
|
|
|
{
|
|
|
|
struct rt6_info *rt = (struct rt6_info *) dst;
|
|
|
|
|
|
|
|
if (rt) {
|
2010-03-28 01:15:45 -06:00
|
|
|
if (rt->rt6i_flags & RTF_CACHE) {
|
|
|
|
if (rt6_check_expired(rt)) {
|
|
|
|
ip6_del_rt(rt);
|
|
|
|
dst = NULL;
|
|
|
|
}
|
|
|
|
} else {
|
2005-04-16 16:20:36 -06:00
|
|
|
dst_release(dst);
|
2010-03-28 01:15:45 -06:00
|
|
|
dst = NULL;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2010-03-28 01:15:45 -06:00
|
|
|
return dst;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static void ip6_link_failure(struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
struct rt6_info *rt;
|
|
|
|
|
2010-02-18 01:25:24 -07:00
|
|
|
icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2009-06-01 23:19:30 -06:00
|
|
|
rt = (struct rt6_info *) skb_dst(skb);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (rt) {
|
2013-07-10 15:00:57 -06:00
|
|
|
if (rt->rt6i_flags & RTF_CACHE) {
|
|
|
|
dst_hold(&rt->dst);
|
|
|
|
if (ip6_del_rt(rt))
|
|
|
|
dst_free(&rt->dst);
|
|
|
|
} else if (rt->rt6i_node && (rt->rt6i_flags & RTF_DEFAULT)) {
|
2005-04-16 16:20:36 -06:00
|
|
|
rt->rt6i_node->fn_sernum = -1;
|
2013-07-10 15:00:57 -06:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-07-17 04:29:28 -06:00
|
|
|
static void ip6_rt_update_pmtu(struct dst_entry *dst, struct sock *sk,
|
|
|
|
struct sk_buff *skb, u32 mtu)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct rt6_info *rt6 = (struct rt6_info*)dst;
|
|
|
|
|
2012-06-15 15:54:11 -06:00
|
|
|
dst_confirm(dst);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (mtu < dst_mtu(dst) && rt6->rt6i_dst.plen == 128) {
|
2012-06-15 15:54:11 -06:00
|
|
|
struct net *net = dev_net(dst->dev);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
rt6->rt6i_flags |= RTF_MODIFIED;
|
|
|
|
if (mtu < IPV6_MIN_MTU) {
|
2010-12-08 22:16:57 -07:00
|
|
|
u32 features = dst_metric(dst, RTAX_FEATURES);
|
2005-04-16 16:20:36 -06:00
|
|
|
mtu = IPV6_MIN_MTU;
|
2010-12-08 22:16:57 -07:00
|
|
|
features |= RTAX_FEATURE_ALLFRAG;
|
|
|
|
dst_metric_set(dst, RTAX_FEATURES, features);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2010-12-08 22:16:57 -07:00
|
|
|
dst_metric_set(dst, RTAX_MTU, mtu);
|
2012-06-15 15:54:11 -06:00
|
|
|
rt6_update_expires(rt6, net->ipv6.sysctl.ip6_rt_mtu_expires);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-06-15 21:01:57 -06:00
|
|
|
void ip6_update_pmtu(struct sk_buff *skb, struct net *net, __be32 mtu,
|
|
|
|
int oif, u32 mark)
|
2012-06-15 15:54:11 -06:00
|
|
|
{
|
|
|
|
const struct ipv6hdr *iph = (struct ipv6hdr *) skb->data;
|
|
|
|
struct dst_entry *dst;
|
|
|
|
struct flowi6 fl6;
|
|
|
|
|
|
|
|
memset(&fl6, 0, sizeof(fl6));
|
|
|
|
fl6.flowi6_oif = oif;
|
|
|
|
fl6.flowi6_mark = mark;
|
|
|
|
fl6.daddr = iph->daddr;
|
|
|
|
fl6.saddr = iph->saddr;
|
2013-01-12 22:01:51 -07:00
|
|
|
fl6.flowlabel = ip6_flowinfo(iph);
|
2012-06-15 15:54:11 -06:00
|
|
|
|
|
|
|
dst = ip6_route_output(net, NULL, &fl6);
|
|
|
|
if (!dst->error)
|
2012-07-17 04:29:28 -06:00
|
|
|
ip6_rt_update_pmtu(dst, NULL, skb, ntohl(mtu));
|
2012-06-15 15:54:11 -06:00
|
|
|
dst_release(dst);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(ip6_update_pmtu);
|
|
|
|
|
|
|
|
void ip6_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, __be32 mtu)
|
|
|
|
{
|
|
|
|
ip6_update_pmtu(skb, sock_net(sk), mtu,
|
|
|
|
sk->sk_bound_dev_if, sk->sk_mark);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(ip6_sk_update_pmtu);
|
|
|
|
|
2013-09-04 05:44:21 -06:00
|
|
|
/* Handle redirects */
|
|
|
|
struct ip6rd_flowi {
|
|
|
|
struct flowi6 fl6;
|
|
|
|
struct in6_addr gateway;
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct rt6_info *__ip6_route_redirect(struct net *net,
|
|
|
|
struct fib6_table *table,
|
|
|
|
struct flowi6 *fl6,
|
|
|
|
int flags)
|
|
|
|
{
|
|
|
|
struct ip6rd_flowi *rdfl = (struct ip6rd_flowi *)fl6;
|
|
|
|
struct rt6_info *rt;
|
|
|
|
struct fib6_node *fn;
|
|
|
|
|
|
|
|
/* Get the "current" route for this destination and
|
|
|
|
* check if the redirect has come from approriate router.
|
|
|
|
*
|
|
|
|
* RFC 4861 specifies that redirects should only be
|
|
|
|
* accepted if they come from the nexthop to the target.
|
|
|
|
* Due to the way the routes are chosen, this notion
|
|
|
|
* is a bit fuzzy and one might need to check all possible
|
|
|
|
* routes.
|
|
|
|
*/
|
|
|
|
|
|
|
|
read_lock_bh(&table->tb6_lock);
|
|
|
|
fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr);
|
|
|
|
restart:
|
|
|
|
for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
|
|
|
|
if (rt6_check_expired(rt))
|
|
|
|
continue;
|
|
|
|
if (rt->dst.error)
|
|
|
|
break;
|
|
|
|
if (!(rt->rt6i_flags & RTF_GATEWAY))
|
|
|
|
continue;
|
|
|
|
if (fl6->flowi6_oif != rt->dst.dev->ifindex)
|
|
|
|
continue;
|
|
|
|
if (!ipv6_addr_equal(&rdfl->gateway, &rt->rt6i_gateway))
|
|
|
|
continue;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!rt)
|
|
|
|
rt = net->ipv6.ip6_null_entry;
|
|
|
|
else if (rt->dst.error) {
|
|
|
|
rt = net->ipv6.ip6_null_entry;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
BACKTRACK(net, &fl6->saddr);
|
|
|
|
out:
|
|
|
|
dst_hold(&rt->dst);
|
|
|
|
|
|
|
|
read_unlock_bh(&table->tb6_lock);
|
|
|
|
|
|
|
|
return rt;
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct dst_entry *ip6_route_redirect(struct net *net,
|
|
|
|
const struct flowi6 *fl6,
|
|
|
|
const struct in6_addr *gateway)
|
|
|
|
{
|
|
|
|
int flags = RT6_LOOKUP_F_HAS_SADDR;
|
|
|
|
struct ip6rd_flowi rdfl;
|
|
|
|
|
|
|
|
rdfl.fl6 = *fl6;
|
|
|
|
rdfl.gateway = *gateway;
|
|
|
|
|
|
|
|
return fib6_rule_lookup(net, &rdfl.fl6,
|
|
|
|
flags, __ip6_route_redirect);
|
|
|
|
}
|
|
|
|
|
2012-07-12 01:08:07 -06:00
|
|
|
void ip6_redirect(struct sk_buff *skb, struct net *net, int oif, u32 mark)
|
|
|
|
{
|
|
|
|
const struct ipv6hdr *iph = (struct ipv6hdr *) skb->data;
|
|
|
|
struct dst_entry *dst;
|
|
|
|
struct flowi6 fl6;
|
|
|
|
|
|
|
|
memset(&fl6, 0, sizeof(fl6));
|
|
|
|
fl6.flowi6_oif = oif;
|
|
|
|
fl6.flowi6_mark = mark;
|
|
|
|
fl6.daddr = iph->daddr;
|
|
|
|
fl6.saddr = iph->saddr;
|
2013-01-12 22:01:51 -07:00
|
|
|
fl6.flowlabel = ip6_flowinfo(iph);
|
2012-07-12 01:08:07 -06:00
|
|
|
|
2013-09-04 05:44:21 -06:00
|
|
|
dst = ip6_route_redirect(net, &fl6, &ipv6_hdr(skb)->saddr);
|
|
|
|
rt6_do_redirect(dst, NULL, skb);
|
2012-07-12 01:08:07 -06:00
|
|
|
dst_release(dst);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(ip6_redirect);
|
|
|
|
|
2013-08-21 22:07:35 -06:00
|
|
|
void ip6_redirect_no_header(struct sk_buff *skb, struct net *net, int oif,
|
|
|
|
u32 mark)
|
|
|
|
{
|
|
|
|
const struct ipv6hdr *iph = ipv6_hdr(skb);
|
|
|
|
const struct rd_msg *msg = (struct rd_msg *)icmp6_hdr(skb);
|
|
|
|
struct dst_entry *dst;
|
|
|
|
struct flowi6 fl6;
|
|
|
|
|
|
|
|
memset(&fl6, 0, sizeof(fl6));
|
|
|
|
fl6.flowi6_oif = oif;
|
|
|
|
fl6.flowi6_mark = mark;
|
|
|
|
fl6.daddr = msg->dest;
|
|
|
|
fl6.saddr = iph->daddr;
|
|
|
|
|
2013-09-04 05:44:21 -06:00
|
|
|
dst = ip6_route_redirect(net, &fl6, &iph->saddr);
|
|
|
|
rt6_do_redirect(dst, NULL, skb);
|
2013-08-21 22:07:35 -06:00
|
|
|
dst_release(dst);
|
|
|
|
}
|
|
|
|
|
2012-07-12 01:08:07 -06:00
|
|
|
void ip6_sk_redirect(struct sk_buff *skb, struct sock *sk)
|
|
|
|
{
|
|
|
|
ip6_redirect(skb, sock_net(sk), sk->sk_bound_dev_if, sk->sk_mark);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(ip6_sk_redirect);
|
|
|
|
|
2010-12-13 13:52:14 -07:00
|
|
|
static unsigned int ip6_default_advmss(const struct dst_entry *dst)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2010-12-13 13:52:14 -07:00
|
|
|
struct net_device *dev = dst->dev;
|
|
|
|
unsigned int mtu = dst_mtu(dst);
|
|
|
|
struct net *net = dev_net(dev);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
mtu -= sizeof(struct ipv6hdr) + sizeof(struct tcphdr);
|
|
|
|
|
2008-03-04 14:47:47 -07:00
|
|
|
if (mtu < net->ipv6.sysctl.ip6_rt_min_advmss)
|
|
|
|
mtu = net->ipv6.sysctl.ip6_rt_min_advmss;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
2007-02-09 07:24:49 -07:00
|
|
|
* Maximal non-jumbo IPv6 payload is IPV6_MAXPLEN and
|
|
|
|
* corresponding MSS is IPV6_MAXPLEN - tcp_header_size.
|
|
|
|
* IPV6_MAXPLEN is also valid and means: "any MSS,
|
2005-04-16 16:20:36 -06:00
|
|
|
* rely only on pmtu discovery"
|
|
|
|
*/
|
|
|
|
if (mtu > IPV6_MAXPLEN - sizeof(struct tcphdr))
|
|
|
|
mtu = IPV6_MAXPLEN;
|
|
|
|
return mtu;
|
|
|
|
}
|
|
|
|
|
2011-11-22 19:12:51 -07:00
|
|
|
static unsigned int ip6_mtu(const struct dst_entry *dst)
|
2010-12-14 14:01:14 -07:00
|
|
|
{
|
|
|
|
struct inet6_dev *idev;
|
2011-11-22 19:13:31 -07:00
|
|
|
unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
|
|
|
|
|
|
|
|
if (mtu)
|
|
|
|
return mtu;
|
|
|
|
|
|
|
|
mtu = IPV6_MIN_MTU;
|
2010-12-14 14:01:14 -07:00
|
|
|
|
|
|
|
rcu_read_lock();
|
|
|
|
idev = __in6_dev_get(dst->dev);
|
|
|
|
if (idev)
|
|
|
|
mtu = idev->cnf.mtu6;
|
|
|
|
rcu_read_unlock();
|
|
|
|
|
|
|
|
return mtu;
|
|
|
|
}
|
|
|
|
|
2007-12-06 17:11:48 -07:00
|
|
|
static struct dst_entry *icmp6_dst_gc_list;
|
|
|
|
static DEFINE_SPINLOCK(icmp6_dst_lock);
|
2006-08-04 04:37:36 -06:00
|
|
|
|
2007-12-06 17:11:48 -07:00
|
|
|
struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
|
2011-12-06 15:04:13 -07:00
|
|
|
struct flowi6 *fl6)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2011-12-06 15:04:13 -07:00
|
|
|
struct dst_entry *dst;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct rt6_info *rt;
|
|
|
|
struct inet6_dev *idev = in6_dev_get(dev);
|
2008-03-25 06:47:49 -06:00
|
|
|
struct net *net = dev_net(dev);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-12-03 16:02:47 -07:00
|
|
|
if (unlikely(!idev))
|
2012-03-14 15:13:11 -06:00
|
|
|
return ERR_PTR(-ENODEV);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2012-06-11 03:01:56 -06:00
|
|
|
rt = ip6_dst_alloc(net, dev, 0, NULL);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (unlikely(!rt)) {
|
2005-04-16 16:20:36 -06:00
|
|
|
in6_dev_put(idev);
|
2011-12-06 15:04:13 -07:00
|
|
|
dst = ERR_PTR(-ENOMEM);
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2011-09-05 15:34:30 -06:00
|
|
|
rt->dst.flags |= DST_HOST;
|
|
|
|
rt->dst.output = ip6_output;
|
2010-06-11 00:31:35 -06:00
|
|
|
atomic_set(&rt->dst.__refcnt, 1);
|
2013-10-20 06:43:04 -06:00
|
|
|
rt->rt6i_gateway = fl6->daddr;
|
2011-12-06 15:04:13 -07:00
|
|
|
rt->rt6i_dst.addr = fl6->daddr;
|
2011-09-05 15:34:30 -06:00
|
|
|
rt->rt6i_dst.plen = 128;
|
|
|
|
rt->rt6i_idev = idev;
|
2012-10-24 00:01:18 -06:00
|
|
|
dst_metric_set(&rt->dst, RTAX_HOPLIMIT, 0);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2007-12-06 17:11:48 -07:00
|
|
|
spin_lock_bh(&icmp6_dst_lock);
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.next = icmp6_dst_gc_list;
|
|
|
|
icmp6_dst_gc_list = &rt->dst;
|
2007-12-06 17:11:48 -07:00
|
|
|
spin_unlock_bh(&icmp6_dst_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-03-04 14:47:47 -07:00
|
|
|
fib6_force_start_gc(net);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-12-06 15:04:13 -07:00
|
|
|
dst = xfrm_lookup(net, &rt->dst, flowi6_to_flowi(fl6), NULL, 0);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
out:
|
2011-12-06 15:04:13 -07:00
|
|
|
return dst;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2008-07-22 15:35:50 -06:00
|
|
|
int icmp6_dst_gc(void)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2011-02-24 22:45:19 -07:00
|
|
|
struct dst_entry *dst, **pprev;
|
2008-07-22 15:35:50 -06:00
|
|
|
int more = 0;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2007-12-06 17:11:48 -07:00
|
|
|
spin_lock_bh(&icmp6_dst_lock);
|
|
|
|
pprev = &icmp6_dst_gc_list;
|
2006-08-04 04:37:36 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
while ((dst = *pprev) != NULL) {
|
|
|
|
if (!atomic_read(&dst->__refcnt)) {
|
|
|
|
*pprev = dst->next;
|
|
|
|
dst_free(dst);
|
|
|
|
} else {
|
|
|
|
pprev = &dst->next;
|
2008-07-22 15:35:50 -06:00
|
|
|
++more;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2007-12-06 17:11:48 -07:00
|
|
|
spin_unlock_bh(&icmp6_dst_lock);
|
2006-08-04 04:37:36 -06:00
|
|
|
|
2008-07-22 15:35:50 -06:00
|
|
|
return more;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2008-09-10 18:27:15 -06:00
|
|
|
static void icmp6_clean_all(int (*func)(struct rt6_info *rt, void *arg),
|
|
|
|
void *arg)
|
|
|
|
{
|
|
|
|
struct dst_entry *dst, **pprev;
|
|
|
|
|
|
|
|
spin_lock_bh(&icmp6_dst_lock);
|
|
|
|
pprev = &icmp6_dst_gc_list;
|
|
|
|
while ((dst = *pprev) != NULL) {
|
|
|
|
struct rt6_info *rt = (struct rt6_info *) dst;
|
|
|
|
if (func(rt, arg)) {
|
|
|
|
*pprev = dst->next;
|
|
|
|
dst_free(dst);
|
|
|
|
} else {
|
|
|
|
pprev = &dst->next;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
spin_unlock_bh(&icmp6_dst_lock);
|
|
|
|
}
|
|
|
|
|
2008-01-18 04:56:57 -07:00
|
|
|
static int ip6_dst_gc(struct dst_ops *ops)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2009-08-28 19:34:49 -06:00
|
|
|
struct net *net = container_of(ops, struct net, ipv6.ip6_dst_ops);
|
2008-03-04 14:50:14 -07:00
|
|
|
int rt_min_interval = net->ipv6.sysctl.ip6_rt_gc_min_interval;
|
|
|
|
int rt_max_size = net->ipv6.sysctl.ip6_rt_max_size;
|
|
|
|
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;
|
2010-10-08 00:37:34 -06:00
|
|
|
int entries;
|
2008-03-04 14:50:14 -07:00
|
|
|
|
2010-10-08 00:37:34 -06:00
|
|
|
entries = dst_entries_get_fast(ops);
|
2013-08-01 02:04:24 -06:00
|
|
|
if (time_after(rt_last_gc + rt_min_interval, jiffies) &&
|
2010-10-08 00:37:34 -06:00
|
|
|
entries <= rt_max_size)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
|
|
|
|
2008-03-04 14:49:47 -07:00
|
|
|
net->ipv6.ip6_rt_gc_expire++;
|
2013-08-01 02:04:14 -06:00
|
|
|
fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net, entries > rt_max_size);
|
2010-10-08 00:37:34 -06:00
|
|
|
entries = dst_entries_get_slow(ops);
|
|
|
|
if (entries < ops->gc_thresh)
|
2008-03-04 14:50:14 -07:00
|
|
|
net->ipv6.ip6_rt_gc_expire = rt_gc_timeout>>1;
|
2005-04-16 16:20:36 -06:00
|
|
|
out:
|
2008-03-04 14:50:14 -07:00
|
|
|
net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>rt_elasticity;
|
2010-10-08 00:37:34 -06:00
|
|
|
return entries > rt_max_size;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
int ip6_route_add(struct fib6_config *cfg)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
int err;
|
2008-03-04 14:47:47 -07:00
|
|
|
struct net *net = cfg->fc_nlinfo.nl_net;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct rt6_info *rt = NULL;
|
|
|
|
struct net_device *dev = NULL;
|
|
|
|
struct inet6_dev *idev = NULL;
|
2006-08-05 00:20:06 -06:00
|
|
|
struct fib6_table *table;
|
2005-04-16 16:20:36 -06:00
|
|
|
int addr_type;
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_dst_len > 128 || cfg->fc_src_len > 128)
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EINVAL;
|
|
|
|
#ifndef CONFIG_IPV6_SUBTREES
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_src_len)
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EINVAL;
|
|
|
|
#endif
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_ifindex) {
|
2005-04-16 16:20:36 -06:00
|
|
|
err = -ENODEV;
|
2008-03-04 14:47:47 -07:00
|
|
|
dev = dev_get_by_index(net, cfg->fc_ifindex);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (!dev)
|
|
|
|
goto out;
|
|
|
|
idev = in6_dev_get(dev);
|
|
|
|
if (!idev)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_metric == 0)
|
|
|
|
cfg->fc_metric = IP6_RT_PRIO_USER;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-11-13 17:14:49 -07:00
|
|
|
err = -ENOBUFS;
|
2011-12-03 16:02:47 -07:00
|
|
|
if (cfg->fc_nlinfo.nlh &&
|
|
|
|
!(cfg->fc_nlinfo.nlh->nlmsg_flags & NLM_F_CREATE)) {
|
2011-11-13 17:14:49 -07:00
|
|
|
table = fib6_get_table(net, cfg->fc_table);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!table) {
|
2012-05-15 08:11:53 -06:00
|
|
|
pr_warn("NLM_F_CREATE should be specified when creating new route\n");
|
2011-11-13 17:14:49 -07:00
|
|
|
table = fib6_new_table(net, cfg->fc_table);
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
table = fib6_new_table(net, cfg->fc_table);
|
|
|
|
}
|
2011-12-03 16:02:47 -07:00
|
|
|
|
|
|
|
if (!table)
|
2006-08-05 00:20:06 -06:00
|
|
|
goto out;
|
|
|
|
|
2012-06-11 03:01:56 -06:00
|
|
|
rt = ip6_dst_alloc(net, NULL, DST_NOCOUNT, table);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!rt) {
|
2005-04-16 16:20:36 -06:00
|
|
|
err = -ENOMEM;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2012-04-05 18:13:10 -06:00
|
|
|
if (cfg->fc_flags & RTF_EXPIRES)
|
|
|
|
rt6_set_expires(rt, jiffies +
|
|
|
|
clock_t_to_jiffies(cfg->fc_expires));
|
|
|
|
else
|
|
|
|
rt6_clean_expires(rt);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_protocol == RTPROT_UNSPEC)
|
|
|
|
cfg->fc_protocol = RTPROT_BOOT;
|
|
|
|
rt->rt6i_protocol = cfg->fc_protocol;
|
|
|
|
|
|
|
|
addr_type = ipv6_addr_type(&cfg->fc_dst);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
if (addr_type & IPV6_ADDR_MULTICAST)
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.input = ip6_mc_input;
|
2010-09-26 18:07:02 -06:00
|
|
|
else if (cfg->fc_flags & RTF_LOCAL)
|
|
|
|
rt->dst.input = ip6_input;
|
2005-04-16 16:20:36 -06:00
|
|
|
else
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.input = ip6_forward;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.output = ip6_output;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
ipv6_addr_prefix(&rt->rt6i_dst.addr, &cfg->fc_dst, cfg->fc_dst_len);
|
|
|
|
rt->rt6i_dst.plen = cfg->fc_dst_len;
|
2005-04-16 16:20:36 -06:00
|
|
|
if (rt->rt6i_dst.plen == 128)
|
2011-06-24 16:23:34 -06:00
|
|
|
rt->dst.flags |= DST_HOST;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-09-05 15:34:30 -06:00
|
|
|
if (!(rt->dst.flags & DST_HOST) && cfg->fc_mx) {
|
|
|
|
u32 *metrics = kzalloc(sizeof(u32) * RTAX_MAX, GFP_KERNEL);
|
|
|
|
if (!metrics) {
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
dst_init_metrics(&rt->dst, metrics, 0);
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
#ifdef CONFIG_IPV6_SUBTREES
|
2006-08-22 01:01:08 -06:00
|
|
|
ipv6_addr_prefix(&rt->rt6i_src.addr, &cfg->fc_src, cfg->fc_src_len);
|
|
|
|
rt->rt6i_src.plen = cfg->fc_src_len;
|
2005-04-16 16:20:36 -06:00
|
|
|
#endif
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
rt->rt6i_metric = cfg->fc_metric;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* We cannot add true routes via loopback here,
|
|
|
|
they would result in kernel looping; promote them to reject routes
|
|
|
|
*/
|
2006-08-22 01:01:08 -06:00
|
|
|
if ((cfg->fc_flags & RTF_REJECT) ||
|
2011-12-03 16:02:47 -07:00
|
|
|
(dev && (dev->flags & IFF_LOOPBACK) &&
|
|
|
|
!(addr_type & IPV6_ADDR_LOOPBACK) &&
|
|
|
|
!(cfg->fc_flags & RTF_LOCAL))) {
|
2005-04-16 16:20:36 -06:00
|
|
|
/* hold loopback dev/idev if we haven't done so. */
|
2008-03-04 14:47:47 -07:00
|
|
|
if (dev != net->loopback_dev) {
|
2005-04-16 16:20:36 -06:00
|
|
|
if (dev) {
|
|
|
|
dev_put(dev);
|
|
|
|
in6_dev_put(idev);
|
|
|
|
}
|
2008-03-04 14:47:47 -07:00
|
|
|
dev = net->loopback_dev;
|
2005-04-16 16:20:36 -06:00
|
|
|
dev_hold(dev);
|
|
|
|
idev = in6_dev_get(dev);
|
|
|
|
if (!idev) {
|
|
|
|
err = -ENODEV;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
rt->rt6i_flags = RTF_REJECT|RTF_NONEXTHOP;
|
2012-09-04 20:12:42 -06:00
|
|
|
switch (cfg->fc_type) {
|
|
|
|
case RTN_BLACKHOLE:
|
|
|
|
rt->dst.error = -EINVAL;
|
2013-12-02 07:25:21 -07:00
|
|
|
rt->dst.output = dst_discard;
|
|
|
|
rt->dst.input = dst_discard;
|
2012-09-04 20:12:42 -06:00
|
|
|
break;
|
|
|
|
case RTN_PROHIBIT:
|
|
|
|
rt->dst.error = -EACCES;
|
2013-12-02 07:25:21 -07:00
|
|
|
rt->dst.output = ip6_pkt_prohibit_out;
|
|
|
|
rt->dst.input = ip6_pkt_prohibit;
|
2012-09-04 20:12:42 -06:00
|
|
|
break;
|
2012-09-05 23:53:35 -06:00
|
|
|
case RTN_THROW:
|
2012-09-04 20:12:42 -06:00
|
|
|
default:
|
2013-12-02 07:25:21 -07:00
|
|
|
rt->dst.error = (cfg->fc_type == RTN_THROW) ? -EAGAIN
|
|
|
|
: -ENETUNREACH;
|
|
|
|
rt->dst.output = ip6_pkt_discard_out;
|
|
|
|
rt->dst.input = ip6_pkt_discard;
|
2012-09-04 20:12:42 -06:00
|
|
|
break;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
goto install_route;
|
|
|
|
}
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_flags & RTF_GATEWAY) {
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct in6_addr *gw_addr;
|
2005-04-16 16:20:36 -06:00
|
|
|
int gwa_type;
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
gw_addr = &cfg->fc_gateway;
|
2011-11-20 20:39:03 -07:00
|
|
|
rt->rt6i_gateway = *gw_addr;
|
2005-04-16 16:20:36 -06:00
|
|
|
gwa_type = ipv6_addr_type(gw_addr);
|
|
|
|
|
|
|
|
if (gwa_type != (IPV6_ADDR_LINKLOCAL|IPV6_ADDR_UNICAST)) {
|
|
|
|
struct rt6_info *grt;
|
|
|
|
|
|
|
|
/* IPv6 strictly inhibits using not link-local
|
|
|
|
addresses as nexthop address.
|
|
|
|
Otherwise, router will not able to send redirects.
|
|
|
|
It is very good, but in some (rare!) circumstances
|
|
|
|
(SIT, PtP, NBMA NOARP links) it is handy to allow
|
|
|
|
some exceptions. --ANK
|
|
|
|
*/
|
|
|
|
err = -EINVAL;
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!(gwa_type & IPV6_ADDR_UNICAST))
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
|
|
|
|
2008-03-04 14:47:47 -07:00
|
|
|
grt = rt6_lookup(net, gw_addr, NULL, cfg->fc_ifindex, 1);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
err = -EHOSTUNREACH;
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!grt)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
|
|
|
if (dev) {
|
2011-12-28 18:19:20 -07:00
|
|
|
if (dev != grt->dst.dev) {
|
2012-10-28 18:13:19 -06:00
|
|
|
ip6_rt_put(grt);
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
} else {
|
2011-12-28 18:19:20 -07:00
|
|
|
dev = grt->dst.dev;
|
2005-04-16 16:20:36 -06:00
|
|
|
idev = grt->rt6i_idev;
|
|
|
|
dev_hold(dev);
|
|
|
|
in6_dev_hold(grt->rt6i_idev);
|
|
|
|
}
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!(grt->rt6i_flags & RTF_GATEWAY))
|
2005-04-16 16:20:36 -06:00
|
|
|
err = 0;
|
2012-10-28 18:13:19 -06:00
|
|
|
ip6_rt_put(grt);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
err = -EINVAL;
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!dev || (dev->flags & IFF_LOOPBACK))
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
err = -ENODEV;
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!dev)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
|
|
|
|
2011-04-13 15:10:57 -06:00
|
|
|
if (!ipv6_addr_any(&cfg->fc_prefsrc)) {
|
|
|
|
if (!ipv6_chk_addr(net, &cfg->fc_prefsrc, dev, 0)) {
|
|
|
|
err = -EINVAL;
|
|
|
|
goto out;
|
|
|
|
}
|
2011-11-20 20:39:03 -07:00
|
|
|
rt->rt6i_prefsrc.addr = cfg->fc_prefsrc;
|
2011-04-13 15:10:57 -06:00
|
|
|
rt->rt6i_prefsrc.plen = 128;
|
|
|
|
} else
|
|
|
|
rt->rt6i_prefsrc.plen = 0;
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
rt->rt6i_flags = cfg->fc_flags;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
install_route:
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_mx) {
|
|
|
|
struct nlattr *nla;
|
|
|
|
int remaining;
|
|
|
|
|
|
|
|
nla_for_each_attr(nla, cfg->fc_mx, cfg->fc_mx_len, remaining) {
|
2007-09-12 06:44:36 -06:00
|
|
|
int type = nla_type(nla);
|
2006-08-22 01:01:08 -06:00
|
|
|
|
|
|
|
if (type) {
|
|
|
|
if (type > RTAX_MAX) {
|
2005-04-16 16:20:36 -06:00
|
|
|
err = -EINVAL;
|
|
|
|
goto out;
|
|
|
|
}
|
2006-08-22 01:01:08 -06:00
|
|
|
|
2010-12-08 22:16:57 -07:00
|
|
|
dst_metric_set(&rt->dst, type, nla_get_u32(nla));
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.dev = dev;
|
2005-04-16 16:20:36 -06:00
|
|
|
rt->rt6i_idev = idev;
|
2006-08-05 00:20:06 -06:00
|
|
|
rt->rt6i_table = table;
|
2008-03-04 00:31:11 -07:00
|
|
|
|
2008-03-25 06:47:49 -06:00
|
|
|
cfg->fc_nlinfo.nl_net = dev_net(dev);
|
2008-03-04 00:31:11 -07:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
return __ip6_ins_rt(rt, &cfg->fc_nlinfo);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
out:
|
|
|
|
if (dev)
|
|
|
|
dev_put(dev);
|
|
|
|
if (idev)
|
|
|
|
in6_dev_put(idev);
|
|
|
|
if (rt)
|
2010-06-11 00:31:35 -06:00
|
|
|
dst_free(&rt->dst);
|
2005-04-16 16:20:36 -06:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
int err;
|
2006-08-05 00:20:06 -06:00
|
|
|
struct fib6_table *table;
|
2011-12-28 18:19:20 -07:00
|
|
|
struct net *net = dev_net(rt->dst.dev);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2012-09-19 13:25:34 -06:00
|
|
|
if (rt == net->ipv6.ip6_null_entry) {
|
|
|
|
err = -ENOENT;
|
|
|
|
goto out;
|
|
|
|
}
|
2006-08-06 23:22:47 -06:00
|
|
|
|
2006-08-05 00:20:06 -06:00
|
|
|
table = rt->rt6i_table;
|
|
|
|
write_lock_bh(&table->tb6_lock);
|
2006-08-22 01:01:08 -06:00
|
|
|
err = fib6_del(rt, info);
|
2006-08-05 00:20:06 -06:00
|
|
|
write_unlock_bh(&table->tb6_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2012-09-19 13:25:34 -06:00
|
|
|
out:
|
2012-10-28 18:13:19 -06:00
|
|
|
ip6_rt_put(rt);
|
2005-04-16 16:20:36 -06:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2006-08-22 01:00:21 -06:00
|
|
|
int ip6_del_rt(struct rt6_info *rt)
|
|
|
|
{
|
2008-01-10 04:26:13 -07:00
|
|
|
struct nl_info info = {
|
2011-12-28 18:19:20 -07:00
|
|
|
.nl_net = dev_net(rt->dst.dev),
|
2008-01-10 04:26:13 -07:00
|
|
|
};
|
2007-12-13 10:45:12 -07:00
|
|
|
return __ip6_del_rt(rt, &info);
|
2006-08-22 01:00:21 -06:00
|
|
|
}
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
static int ip6_route_del(struct fib6_config *cfg)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-08-05 00:20:06 -06:00
|
|
|
struct fib6_table *table;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct fib6_node *fn;
|
|
|
|
struct rt6_info *rt;
|
|
|
|
int err = -ESRCH;
|
|
|
|
|
2008-03-04 14:47:47 -07:00
|
|
|
table = fib6_get_table(cfg->fc_nlinfo.nl_net, cfg->fc_table);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!table)
|
2006-08-05 00:20:06 -06:00
|
|
|
return err;
|
|
|
|
|
|
|
|
read_lock_bh(&table->tb6_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-05 00:20:06 -06:00
|
|
|
fn = fib6_locate(&table->tb6_root,
|
2006-08-22 01:01:08 -06:00
|
|
|
&cfg->fc_dst, cfg->fc_dst_len,
|
|
|
|
&cfg->fc_src, cfg->fc_src_len);
|
2007-02-09 07:24:49 -07:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
if (fn) {
|
2010-06-11 00:31:35 -06:00
|
|
|
for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_ifindex &&
|
2011-12-28 18:19:20 -07:00
|
|
|
(!rt->dst.dev ||
|
|
|
|
rt->dst.dev->ifindex != cfg->fc_ifindex))
|
2005-04-16 16:20:36 -06:00
|
|
|
continue;
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_flags & RTF_GATEWAY &&
|
|
|
|
!ipv6_addr_equal(&cfg->fc_gateway, &rt->rt6i_gateway))
|
2005-04-16 16:20:36 -06:00
|
|
|
continue;
|
2006-08-22 01:01:08 -06:00
|
|
|
if (cfg->fc_metric && cfg->fc_metric != rt->rt6i_metric)
|
2005-04-16 16:20:36 -06:00
|
|
|
continue;
|
2010-06-11 00:31:35 -06:00
|
|
|
dst_hold(&rt->dst);
|
2006-08-05 00:20:06 -06:00
|
|
|
read_unlock_bh(&table->tb6_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
return __ip6_del_rt(rt, &cfg->fc_nlinfo);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
}
|
2006-08-05 00:20:06 -06:00
|
|
|
read_unlock_bh(&table->tb6_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2012-07-17 04:29:28 -06:00
|
|
|
static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_buff *skb)
|
2006-08-23 18:18:26 -06:00
|
|
|
{
|
2012-07-12 00:43:53 -06:00
|
|
|
struct net *net = dev_net(skb->dev);
|
2006-08-23 18:18:26 -06:00
|
|
|
struct netevent_redirect netevent;
|
2012-07-12 00:43:53 -06:00
|
|
|
struct rt6_info *rt, *nrt = NULL;
|
|
|
|
struct ndisc_options ndopts;
|
|
|
|
struct inet6_dev *in6_dev;
|
|
|
|
struct neighbour *neigh;
|
2013-01-05 09:34:51 -07:00
|
|
|
struct rd_msg *msg;
|
2012-07-12 01:05:02 -06:00
|
|
|
int optlen, on_link;
|
|
|
|
u8 *lladdr;
|
2012-07-12 00:43:53 -06:00
|
|
|
|
2013-05-28 14:34:26 -06:00
|
|
|
optlen = skb_tail_pointer(skb) - skb_transport_header(skb);
|
2013-01-05 09:34:51 -07:00
|
|
|
optlen -= sizeof(*msg);
|
2012-07-12 00:43:53 -06:00
|
|
|
|
|
|
|
if (optlen < 0) {
|
2012-07-12 01:05:02 -06:00
|
|
|
net_dbg_ratelimited("rt6_do_redirect: packet too short\n");
|
2012-07-12 00:43:53 -06:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2013-01-05 09:34:51 -07:00
|
|
|
msg = (struct rd_msg *)icmp6_hdr(skb);
|
2012-07-12 00:43:53 -06:00
|
|
|
|
2013-01-05 09:34:51 -07:00
|
|
|
if (ipv6_addr_is_multicast(&msg->dest)) {
|
2012-07-12 01:05:02 -06:00
|
|
|
net_dbg_ratelimited("rt6_do_redirect: destination address is multicast\n");
|
2012-07-12 00:43:53 -06:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2012-07-12 01:05:02 -06:00
|
|
|
on_link = 0;
|
2013-01-05 09:34:51 -07:00
|
|
|
if (ipv6_addr_equal(&msg->dest, &msg->target)) {
|
2012-07-12 00:43:53 -06:00
|
|
|
on_link = 1;
|
2013-01-05 09:34:51 -07:00
|
|
|
} else if (ipv6_addr_type(&msg->target) !=
|
2012-07-12 00:43:53 -06:00
|
|
|
(IPV6_ADDR_UNICAST|IPV6_ADDR_LINKLOCAL)) {
|
2012-07-12 01:05:02 -06:00
|
|
|
net_dbg_ratelimited("rt6_do_redirect: target address is not link-local unicast\n");
|
2012-07-12 00:43:53 -06:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
in6_dev = __in6_dev_get(skb->dev);
|
|
|
|
if (!in6_dev)
|
|
|
|
return;
|
|
|
|
if (in6_dev->cnf.forwarding || !in6_dev->cnf.accept_redirects)
|
|
|
|
return;
|
|
|
|
|
|
|
|
/* RFC2461 8.1:
|
|
|
|
* The IP source address of the Redirect MUST be the same as the current
|
|
|
|
* first-hop router for the specified ICMP Destination Address.
|
|
|
|
*/
|
|
|
|
|
2013-01-05 09:34:51 -07:00
|
|
|
if (!ndisc_parse_options(msg->opt, optlen, &ndopts)) {
|
2012-07-12 00:43:53 -06:00
|
|
|
net_dbg_ratelimited("rt6_redirect: invalid ND options\n");
|
|
|
|
return;
|
|
|
|
}
|
2012-07-12 01:05:02 -06:00
|
|
|
|
|
|
|
lladdr = NULL;
|
2012-07-12 00:43:53 -06:00
|
|
|
if (ndopts.nd_opts_tgt_lladdr) {
|
|
|
|
lladdr = ndisc_opt_addr_data(ndopts.nd_opts_tgt_lladdr,
|
|
|
|
skb->dev);
|
|
|
|
if (!lladdr) {
|
|
|
|
net_dbg_ratelimited("rt6_redirect: invalid link-layer address length\n");
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-07-12 01:05:02 -06:00
|
|
|
rt = (struct rt6_info *) dst;
|
|
|
|
if (rt == net->ipv6.ip6_null_entry) {
|
|
|
|
net_dbg_ratelimited("rt6_redirect: source isn't a valid nexthop for redirect target\n");
|
2012-07-12 00:43:53 -06:00
|
|
|
return;
|
2012-07-12 01:05:02 -06:00
|
|
|
}
|
2012-07-12 00:43:53 -06:00
|
|
|
|
2012-07-12 01:05:02 -06:00
|
|
|
/* 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->dst);
|
2006-08-23 18:18:26 -06:00
|
|
|
|
2013-01-05 09:34:51 -07:00
|
|
|
neigh = __neigh_lookup(&nd_tbl, &msg->target, skb->dev, 1);
|
2012-07-12 01:05:02 -06:00
|
|
|
if (!neigh)
|
|
|
|
return;
|
2006-08-23 18:18:26 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* We have finally decided to accept it.
|
|
|
|
*/
|
|
|
|
|
2007-02-09 07:24:49 -07:00
|
|
|
neigh_update(neigh, lladdr, NUD_STALE,
|
2005-04-16 16:20:36 -06:00
|
|
|
NEIGH_UPDATE_F_WEAK_OVERRIDE|
|
|
|
|
NEIGH_UPDATE_F_OVERRIDE|
|
|
|
|
(on_link ? 0 : (NEIGH_UPDATE_F_OVERRIDE_ISROUTER|
|
|
|
|
NEIGH_UPDATE_F_ISROUTER))
|
|
|
|
);
|
|
|
|
|
2013-01-05 09:34:51 -07:00
|
|
|
nrt = ip6_rt_copy(rt, &msg->dest);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!nrt)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
nrt->rt6i_flags = RTF_GATEWAY|RTF_UP|RTF_DYNAMIC|RTF_CACHE;
|
|
|
|
if (on_link)
|
|
|
|
nrt->rt6i_flags &= ~RTF_GATEWAY;
|
|
|
|
|
2011-11-20 20:39:03 -07:00
|
|
|
nrt->rt6i_gateway = *(struct in6_addr *)neigh->primary_key;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:00:45 -06:00
|
|
|
if (ip6_ins_rt(nrt))
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
|
|
|
|
2010-06-11 00:31:35 -06:00
|
|
|
netevent.old = &rt->dst;
|
|
|
|
netevent.new = &nrt->dst;
|
2013-01-05 09:34:51 -07:00
|
|
|
netevent.daddr = &msg->dest;
|
2013-01-14 02:28:27 -07:00
|
|
|
netevent.neigh = neigh;
|
2006-07-30 21:43:36 -06:00
|
|
|
call_netevent_notifiers(NETEVENT_REDIRECT, &netevent);
|
|
|
|
|
2011-12-03 16:02:47 -07:00
|
|
|
if (rt->rt6i_flags & RTF_CACHE) {
|
2012-07-12 01:05:02 -06:00
|
|
|
rt = (struct rt6_info *) dst_clone(&rt->dst);
|
2006-08-22 01:00:21 -06:00
|
|
|
ip6_del_rt(rt);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
out:
|
2012-07-12 00:43:53 -06:00
|
|
|
neigh_release(neigh);
|
2012-07-12 01:05:02 -06:00
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Misc support functions
|
|
|
|
*/
|
|
|
|
|
2012-04-05 18:13:10 -06:00
|
|
|
static struct rt6_info *ip6_rt_copy(struct rt6_info *ort,
|
2011-07-19 14:18:36 -06:00
|
|
|
const struct in6_addr *dest)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2011-12-28 18:19:20 -07:00
|
|
|
struct net *net = dev_net(ort->dst.dev);
|
2012-06-11 03:01:56 -06:00
|
|
|
struct rt6_info *rt = ip6_dst_alloc(net, ort->dst.dev, 0,
|
|
|
|
ort->rt6i_table);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
if (rt) {
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.input = ort->dst.input;
|
|
|
|
rt->dst.output = ort->dst.output;
|
2011-09-05 15:34:30 -06:00
|
|
|
rt->dst.flags |= DST_HOST;
|
2010-06-11 00:31:35 -06:00
|
|
|
|
2011-11-20 20:39:03 -07:00
|
|
|
rt->rt6i_dst.addr = *dest;
|
2011-09-05 15:34:30 -06:00
|
|
|
rt->rt6i_dst.plen = 128;
|
2010-12-08 22:16:57 -07:00
|
|
|
dst_copy_metrics(&rt->dst, &ort->dst);
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.error = ort->dst.error;
|
2005-04-16 16:20:36 -06:00
|
|
|
rt->rt6i_idev = ort->rt6i_idev;
|
|
|
|
if (rt->rt6i_idev)
|
|
|
|
in6_dev_hold(rt->rt6i_idev);
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.lastuse = jiffies;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2013-10-20 06:43:04 -06:00
|
|
|
if (ort->rt6i_flags & RTF_GATEWAY)
|
|
|
|
rt->rt6i_gateway = ort->rt6i_gateway;
|
|
|
|
else
|
|
|
|
rt->rt6i_gateway = *dest;
|
2012-04-05 18:13:10 -06:00
|
|
|
rt->rt6i_flags = ort->rt6i_flags;
|
2013-12-18 21:40:26 -07:00
|
|
|
rt6_set_from(rt, ort);
|
2005-04-16 16:20:36 -06:00
|
|
|
rt->rt6i_metric = 0;
|
|
|
|
|
|
|
|
#ifdef CONFIG_IPV6_SUBTREES
|
|
|
|
memcpy(&rt->rt6i_src, &ort->rt6i_src, sizeof(struct rt6key));
|
|
|
|
#endif
|
2011-05-20 05:27:24 -06:00
|
|
|
memcpy(&rt->rt6i_prefsrc, &ort->rt6i_prefsrc, sizeof(struct rt6key));
|
2006-08-05 00:20:06 -06:00
|
|
|
rt->rt6i_table = ort->rt6i_table;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
return rt;
|
|
|
|
}
|
|
|
|
|
2006-03-20 18:06:24 -07:00
|
|
|
#ifdef CONFIG_IPV6_ROUTE_INFO
|
2008-03-04 14:46:48 -07:00
|
|
|
static struct rt6_info *rt6_get_route_info(struct net *net,
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct in6_addr *prefix, int prefixlen,
|
|
|
|
const struct in6_addr *gwaddr, int ifindex)
|
2006-03-20 18:06:24 -07:00
|
|
|
{
|
|
|
|
struct fib6_node *fn;
|
|
|
|
struct rt6_info *rt = NULL;
|
2006-08-05 00:20:06 -06:00
|
|
|
struct fib6_table *table;
|
|
|
|
|
2008-03-04 14:46:48 -07:00
|
|
|
table = fib6_get_table(net, RT6_TABLE_INFO);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!table)
|
2006-08-05 00:20:06 -06:00
|
|
|
return NULL;
|
2006-03-20 18:06:24 -07:00
|
|
|
|
2012-09-11 15:59:01 -06:00
|
|
|
read_lock_bh(&table->tb6_lock);
|
2006-08-05 00:20:06 -06:00
|
|
|
fn = fib6_locate(&table->tb6_root, prefix ,prefixlen, NULL, 0);
|
2006-03-20 18:06:24 -07:00
|
|
|
if (!fn)
|
|
|
|
goto out;
|
|
|
|
|
2010-06-11 00:31:35 -06:00
|
|
|
for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
|
2011-12-28 18:19:20 -07:00
|
|
|
if (rt->dst.dev->ifindex != ifindex)
|
2006-03-20 18:06:24 -07:00
|
|
|
continue;
|
|
|
|
if ((rt->rt6i_flags & (RTF_ROUTEINFO|RTF_GATEWAY)) != (RTF_ROUTEINFO|RTF_GATEWAY))
|
|
|
|
continue;
|
|
|
|
if (!ipv6_addr_equal(&rt->rt6i_gateway, gwaddr))
|
|
|
|
continue;
|
2010-06-11 00:31:35 -06:00
|
|
|
dst_hold(&rt->dst);
|
2006-03-20 18:06:24 -07:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
out:
|
2012-09-11 15:59:01 -06:00
|
|
|
read_unlock_bh(&table->tb6_lock);
|
2006-03-20 18:06:24 -07:00
|
|
|
return rt;
|
|
|
|
}
|
|
|
|
|
2008-03-04 14:46:48 -07:00
|
|
|
static struct rt6_info *rt6_add_route_info(struct net *net,
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct in6_addr *prefix, int prefixlen,
|
|
|
|
const struct in6_addr *gwaddr, int ifindex,
|
2012-04-14 23:58:06 -06:00
|
|
|
unsigned int pref)
|
2006-03-20 18:06:24 -07:00
|
|
|
{
|
2006-08-22 01:01:08 -06:00
|
|
|
struct fib6_config cfg = {
|
|
|
|
.fc_table = RT6_TABLE_INFO,
|
2008-02-10 00:43:11 -07:00
|
|
|
.fc_metric = IP6_RT_PRIO_USER,
|
2006-08-22 01:01:08 -06:00
|
|
|
.fc_ifindex = ifindex,
|
|
|
|
.fc_dst_len = prefixlen,
|
|
|
|
.fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO |
|
|
|
|
RTF_UP | RTF_PREF(pref),
|
2012-09-07 14:12:54 -06:00
|
|
|
.fc_nlinfo.portid = 0,
|
2008-03-04 14:46:48 -07:00
|
|
|
.fc_nlinfo.nlh = NULL,
|
|
|
|
.fc_nlinfo.nl_net = net,
|
2006-08-22 01:01:08 -06:00
|
|
|
};
|
|
|
|
|
2011-11-20 20:39:03 -07:00
|
|
|
cfg.fc_dst = *prefix;
|
|
|
|
cfg.fc_gateway = *gwaddr;
|
2006-03-20 18:06:24 -07:00
|
|
|
|
2006-03-20 18:06:42 -07:00
|
|
|
/* We should treat it as a default route if prefix length is 0. */
|
|
|
|
if (!prefixlen)
|
2006-08-22 01:01:08 -06:00
|
|
|
cfg.fc_flags |= RTF_DEFAULT;
|
2006-03-20 18:06:24 -07:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
ip6_route_add(&cfg);
|
2006-03-20 18:06:24 -07:00
|
|
|
|
2008-03-04 14:46:48 -07:00
|
|
|
return rt6_get_route_info(net, prefix, prefixlen, gwaddr, ifindex);
|
2006-03-20 18:06:24 -07:00
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2011-04-21 22:53:02 -06:00
|
|
|
struct rt6_info *rt6_get_dflt_router(const struct in6_addr *addr, struct net_device *dev)
|
2007-02-09 07:24:49 -07:00
|
|
|
{
|
2005-04-16 16:20:36 -06:00
|
|
|
struct rt6_info *rt;
|
2006-08-05 00:20:06 -06:00
|
|
|
struct fib6_table *table;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-03-25 06:47:49 -06:00
|
|
|
table = fib6_get_table(dev_net(dev), RT6_TABLE_DFLT);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!table)
|
2006-08-05 00:20:06 -06:00
|
|
|
return NULL;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2012-09-11 15:59:01 -06:00
|
|
|
read_lock_bh(&table->tb6_lock);
|
2010-06-11 00:31:35 -06:00
|
|
|
for (rt = table->tb6_root.leaf; rt; rt=rt->dst.rt6_next) {
|
2011-12-28 18:19:20 -07:00
|
|
|
if (dev == rt->dst.dev &&
|
2006-03-20 18:00:48 -07:00
|
|
|
((rt->rt6i_flags & (RTF_ADDRCONF | RTF_DEFAULT)) == (RTF_ADDRCONF | RTF_DEFAULT)) &&
|
2005-04-16 16:20:36 -06:00
|
|
|
ipv6_addr_equal(&rt->rt6i_gateway, addr))
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (rt)
|
2010-06-11 00:31:35 -06:00
|
|
|
dst_hold(&rt->dst);
|
2012-09-11 15:59:01 -06:00
|
|
|
read_unlock_bh(&table->tb6_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
return rt;
|
|
|
|
}
|
|
|
|
|
2011-04-21 22:53:02 -06:00
|
|
|
struct rt6_info *rt6_add_dflt_router(const struct in6_addr *gwaddr,
|
2006-03-20 18:04:53 -07:00
|
|
|
struct net_device *dev,
|
|
|
|
unsigned int pref)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-08-22 01:01:08 -06:00
|
|
|
struct fib6_config cfg = {
|
|
|
|
.fc_table = RT6_TABLE_DFLT,
|
2008-02-10 00:43:11 -07:00
|
|
|
.fc_metric = IP6_RT_PRIO_USER,
|
2006-08-22 01:01:08 -06:00
|
|
|
.fc_ifindex = dev->ifindex,
|
|
|
|
.fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT |
|
|
|
|
RTF_UP | RTF_EXPIRES | RTF_PREF(pref),
|
2012-09-07 14:12:54 -06:00
|
|
|
.fc_nlinfo.portid = 0,
|
2008-03-04 14:47:47 -07:00
|
|
|
.fc_nlinfo.nlh = NULL,
|
2008-03-25 06:47:49 -06:00
|
|
|
.fc_nlinfo.nl_net = dev_net(dev),
|
2006-08-22 01:01:08 -06:00
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-11-20 20:39:03 -07:00
|
|
|
cfg.fc_gateway = *gwaddr;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
ip6_route_add(&cfg);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
return rt6_get_dflt_router(gwaddr, dev);
|
|
|
|
}
|
|
|
|
|
2008-03-04 14:47:14 -07:00
|
|
|
void rt6_purge_dflt_routers(struct net *net)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct rt6_info *rt;
|
2006-08-05 00:20:06 -06:00
|
|
|
struct fib6_table *table;
|
|
|
|
|
|
|
|
/* NOTE: Keep consistent with rt6_get_dflt_router */
|
2008-03-04 14:47:14 -07:00
|
|
|
table = fib6_get_table(net, RT6_TABLE_DFLT);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!table)
|
2006-08-05 00:20:06 -06:00
|
|
|
return;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
restart:
|
2006-08-05 00:20:06 -06:00
|
|
|
read_lock_bh(&table->tb6_lock);
|
2010-06-11 00:31:35 -06:00
|
|
|
for (rt = table->tb6_root.leaf; rt; rt = rt->dst.rt6_next) {
|
2013-03-03 13:46:46 -07:00
|
|
|
if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF) &&
|
|
|
|
(!rt->rt6i_idev || rt->rt6i_idev->cnf.accept_ra != 2)) {
|
2010-06-11 00:31:35 -06:00
|
|
|
dst_hold(&rt->dst);
|
2006-08-05 00:20:06 -06:00
|
|
|
read_unlock_bh(&table->tb6_lock);
|
2006-08-22 01:00:21 -06:00
|
|
|
ip6_del_rt(rt);
|
2005-04-16 16:20:36 -06:00
|
|
|
goto restart;
|
|
|
|
}
|
|
|
|
}
|
2006-08-05 00:20:06 -06:00
|
|
|
read_unlock_bh(&table->tb6_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2008-03-04 14:47:47 -07:00
|
|
|
static void rtmsg_to_fib6_config(struct net *net,
|
|
|
|
struct in6_rtmsg *rtmsg,
|
2006-08-22 01:01:08 -06:00
|
|
|
struct fib6_config *cfg)
|
|
|
|
{
|
|
|
|
memset(cfg, 0, sizeof(*cfg));
|
|
|
|
|
|
|
|
cfg->fc_table = RT6_TABLE_MAIN;
|
|
|
|
cfg->fc_ifindex = rtmsg->rtmsg_ifindex;
|
|
|
|
cfg->fc_metric = rtmsg->rtmsg_metric;
|
|
|
|
cfg->fc_expires = rtmsg->rtmsg_info;
|
|
|
|
cfg->fc_dst_len = rtmsg->rtmsg_dst_len;
|
|
|
|
cfg->fc_src_len = rtmsg->rtmsg_src_len;
|
|
|
|
cfg->fc_flags = rtmsg->rtmsg_flags;
|
|
|
|
|
2008-03-04 14:47:47 -07:00
|
|
|
cfg->fc_nlinfo.nl_net = net;
|
2008-02-26 19:10:03 -07:00
|
|
|
|
2011-11-20 20:39:03 -07:00
|
|
|
cfg->fc_dst = rtmsg->rtmsg_dst;
|
|
|
|
cfg->fc_src = rtmsg->rtmsg_src;
|
|
|
|
cfg->fc_gateway = rtmsg->rtmsg_gateway;
|
2006-08-22 01:01:08 -06:00
|
|
|
}
|
|
|
|
|
2008-03-04 14:47:47 -07:00
|
|
|
int ipv6_route_ioctl(struct net *net, unsigned int cmd, void __user *arg)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-08-22 01:01:08 -06:00
|
|
|
struct fib6_config cfg;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct in6_rtmsg rtmsg;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
switch(cmd) {
|
|
|
|
case SIOCADDRT: /* Add a route */
|
|
|
|
case SIOCDELRT: /* Delete a route */
|
net: Allow userns root to control ipv6
Allow an unpriviled user who has created a user namespace, and then
created a network namespace to effectively use the new network
namespace, by reducing capable(CAP_NET_ADMIN) and
capable(CAP_NET_RAW) calls to be ns_capable(net->user_ns,
CAP_NET_ADMIN), or capable(net->user_ns, CAP_NET_RAW) calls.
Settings that merely control a single network device are allowed.
Either the network device is a logical network device where
restrictions make no difference or the network device is hardware NIC
that has been explicity moved from the initial network namespace.
In general policy and network stack state changes are allowed while
resource control is left unchanged.
Allow the SIOCSIFADDR ioctl to add ipv6 addresses.
Allow the SIOCDIFADDR ioctl to delete ipv6 addresses.
Allow the SIOCADDRT ioctl to add ipv6 routes.
Allow the SIOCDELRT ioctl to delete ipv6 routes.
Allow creation of ipv6 raw sockets.
Allow setting the IPV6_JOIN_ANYCAST socket option.
Allow setting the IPV6_FL_A_RENEW parameter of the IPV6_FLOWLABEL_MGR
socket option.
Allow setting the IPV6_TRANSPARENT socket option.
Allow setting the IPV6_HOPOPTS socket option.
Allow setting the IPV6_RTHDRDSTOPTS socket option.
Allow setting the IPV6_DSTOPTS socket option.
Allow setting the IPV6_IPSEC_POLICY socket option.
Allow setting the IPV6_XFRM_POLICY socket option.
Allow sending packets with the IPV6_2292HOPOPTS control message.
Allow sending packets with the IPV6_2292DSTOPTS control message.
Allow sending packets with the IPV6_RTHDRDSTOPTS control message.
Allow setting the multicast routing socket options on non multicast
routing sockets.
Allow the SIOCADDTUNNEL, SIOCCHGTUNNEL, and SIOCDELTUNNEL ioctls for
setting up, changing and deleting tunnels over ipv6.
Allow the SIOCADDTUNNEL, SIOCCHGTUNNEL, SIOCDELTUNNEL ioctls for
setting up, changing and deleting ipv6 over ipv4 tunnels.
Allow the SIOCADDPRL, SIOCDELPRL, SIOCCHGPRL ioctls for adding,
deleting, and changing the potential router list for ISATAP tunnels.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-15 20:03:06 -07:00
|
|
|
if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EPERM;
|
|
|
|
err = copy_from_user(&rtmsg, arg,
|
|
|
|
sizeof(struct in6_rtmsg));
|
|
|
|
if (err)
|
|
|
|
return -EFAULT;
|
2006-08-22 01:01:08 -06:00
|
|
|
|
2008-03-04 14:47:47 -07:00
|
|
|
rtmsg_to_fib6_config(net, &rtmsg, &cfg);
|
2006-08-22 01:01:08 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
rtnl_lock();
|
|
|
|
switch (cmd) {
|
|
|
|
case SIOCADDRT:
|
2006-08-22 01:01:08 -06:00
|
|
|
err = ip6_route_add(&cfg);
|
2005-04-16 16:20:36 -06:00
|
|
|
break;
|
|
|
|
case SIOCDELRT:
|
2006-08-22 01:01:08 -06:00
|
|
|
err = ip6_route_del(&cfg);
|
2005-04-16 16:20:36 -06:00
|
|
|
break;
|
|
|
|
default:
|
|
|
|
err = -EINVAL;
|
|
|
|
}
|
|
|
|
rtnl_unlock();
|
|
|
|
|
|
|
|
return err;
|
2007-04-20 18:09:22 -06:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Drop the packet on the floor
|
|
|
|
*/
|
|
|
|
|
2009-06-23 05:31:07 -06:00
|
|
|
static int ip6_pkt_drop(struct sk_buff *skb, u8 code, int ipstats_mib_noroutes)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2007-04-13 17:18:02 -06:00
|
|
|
int type;
|
2009-06-01 23:19:30 -06:00
|
|
|
struct dst_entry *dst = skb_dst(skb);
|
2007-04-13 17:18:02 -06:00
|
|
|
switch (ipstats_mib_noroutes) {
|
|
|
|
case IPSTATS_MIB_INNOROUTES:
|
2007-04-25 18:54:47 -06:00
|
|
|
type = ipv6_addr_type(&ipv6_hdr(skb)->daddr);
|
2010-02-25 16:28:58 -07:00
|
|
|
if (type == IPV6_ADDR_ANY) {
|
2008-10-08 11:54:51 -06:00
|
|
|
IP6_INC_STATS(dev_net(dst->dev), ip6_dst_idev(dst),
|
|
|
|
IPSTATS_MIB_INADDRERRORS);
|
2007-04-13 17:18:02 -06:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
/* FALLTHROUGH */
|
|
|
|
case IPSTATS_MIB_OUTNOROUTES:
|
2008-10-08 11:54:51 -06:00
|
|
|
IP6_INC_STATS(dev_net(dst->dev), ip6_dst_idev(dst),
|
|
|
|
ipstats_mib_noroutes);
|
2007-04-13 17:18:02 -06:00
|
|
|
break;
|
|
|
|
}
|
2010-02-18 01:25:24 -07:00
|
|
|
icmpv6_send(skb, ICMPV6_DEST_UNREACH, code, 0);
|
2005-04-16 16:20:36 -06:00
|
|
|
kfree_skb(skb);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2006-10-18 21:46:54 -06:00
|
|
|
static int ip6_pkt_discard(struct sk_buff *skb)
|
|
|
|
{
|
2007-04-13 17:18:02 -06:00
|
|
|
return ip6_pkt_drop(skb, ICMPV6_NOROUTE, IPSTATS_MIB_INNOROUTES);
|
2006-10-18 21:46:54 -06:00
|
|
|
}
|
|
|
|
|
2005-08-15 23:18:02 -06:00
|
|
|
static int ip6_pkt_discard_out(struct sk_buff *skb)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2009-06-01 23:19:30 -06:00
|
|
|
skb->dev = skb_dst(skb)->dev;
|
2007-04-13 17:18:02 -06:00
|
|
|
return ip6_pkt_drop(skb, ICMPV6_NOROUTE, IPSTATS_MIB_OUTNOROUTES);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-10-18 21:46:54 -06:00
|
|
|
static int ip6_pkt_prohibit(struct sk_buff *skb)
|
|
|
|
{
|
2007-04-13 17:18:02 -06:00
|
|
|
return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED, IPSTATS_MIB_INNOROUTES);
|
2006-10-18 21:46:54 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ip6_pkt_prohibit_out(struct sk_buff *skb)
|
|
|
|
{
|
2009-06-01 23:19:30 -06:00
|
|
|
skb->dev = skb_dst(skb)->dev;
|
2007-04-13 17:18:02 -06:00
|
|
|
return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED, IPSTATS_MIB_OUTNOROUTES);
|
2006-10-18 21:46:54 -06:00
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Allocate a dst for local (unicast / anycast) address.
|
|
|
|
*/
|
|
|
|
|
|
|
|
struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
|
|
|
|
const struct in6_addr *addr,
|
2011-12-06 14:48:14 -07:00
|
|
|
bool anycast)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2008-03-25 06:47:49 -06:00
|
|
|
struct net *net = dev_net(idev->dev);
|
2013-12-06 19:33:45 -07:00
|
|
|
struct rt6_info *rt = ip6_dst_alloc(net, net->loopback_dev,
|
|
|
|
DST_NOCOUNT, NULL);
|
|
|
|
if (!rt)
|
2005-04-16 16:20:36 -06:00
|
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
|
|
|
|
in6_dev_hold(idev);
|
|
|
|
|
2011-06-24 16:23:34 -06:00
|
|
|
rt->dst.flags |= DST_HOST;
|
2010-06-11 00:31:35 -06:00
|
|
|
rt->dst.input = ip6_input;
|
|
|
|
rt->dst.output = ip6_output;
|
2005-04-16 16:20:36 -06:00
|
|
|
rt->rt6i_idev = idev;
|
|
|
|
|
|
|
|
rt->rt6i_flags = RTF_UP | RTF_NONEXTHOP;
|
2005-12-21 06:56:42 -07:00
|
|
|
if (anycast)
|
|
|
|
rt->rt6i_flags |= RTF_ANYCAST;
|
|
|
|
else
|
2005-04-16 16:20:36 -06:00
|
|
|
rt->rt6i_flags |= RTF_LOCAL;
|
|
|
|
|
2013-10-20 06:43:04 -06:00
|
|
|
rt->rt6i_gateway = *addr;
|
2011-11-20 20:39:03 -07:00
|
|
|
rt->rt6i_dst.addr = *addr;
|
2005-04-16 16:20:36 -06:00
|
|
|
rt->rt6i_dst.plen = 128;
|
2008-03-04 14:47:47 -07:00
|
|
|
rt->rt6i_table = fib6_get_table(net, RT6_TABLE_LOCAL);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-06-11 00:31:35 -06:00
|
|
|
atomic_set(&rt->dst.__refcnt, 1);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
return rt;
|
|
|
|
}
|
|
|
|
|
2011-04-13 15:10:57 -06:00
|
|
|
int ip6_route_get_saddr(struct net *net,
|
|
|
|
struct rt6_info *rt,
|
2011-04-21 22:53:02 -06:00
|
|
|
const struct in6_addr *daddr,
|
2011-04-13 15:10:57 -06:00
|
|
|
unsigned int prefs,
|
|
|
|
struct in6_addr *saddr)
|
|
|
|
{
|
|
|
|
struct inet6_dev *idev = ip6_dst_idev((struct dst_entry*)rt);
|
|
|
|
int err = 0;
|
|
|
|
if (rt->rt6i_prefsrc.plen)
|
2011-11-20 20:39:03 -07:00
|
|
|
*saddr = rt->rt6i_prefsrc.addr;
|
2011-04-13 15:10:57 -06:00
|
|
|
else
|
|
|
|
err = ipv6_dev_get_saddr(net, idev ? idev->dev : NULL,
|
|
|
|
daddr, prefs, saddr);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* remove deleted ip from prefsrc entries */
|
|
|
|
struct arg_dev_net_ip {
|
|
|
|
struct net_device *dev;
|
|
|
|
struct net *net;
|
|
|
|
struct in6_addr *addr;
|
|
|
|
};
|
|
|
|
|
|
|
|
static int fib6_remove_prefsrc(struct rt6_info *rt, void *arg)
|
|
|
|
{
|
|
|
|
struct net_device *dev = ((struct arg_dev_net_ip *)arg)->dev;
|
|
|
|
struct net *net = ((struct arg_dev_net_ip *)arg)->net;
|
|
|
|
struct in6_addr *addr = ((struct arg_dev_net_ip *)arg)->addr;
|
|
|
|
|
2011-12-28 18:19:20 -07:00
|
|
|
if (((void *)rt->dst.dev == dev || !dev) &&
|
2011-04-13 15:10:57 -06:00
|
|
|
rt != net->ipv6.ip6_null_entry &&
|
|
|
|
ipv6_addr_equal(addr, &rt->rt6i_prefsrc.addr)) {
|
|
|
|
/* remove prefsrc entry */
|
|
|
|
rt->rt6i_prefsrc.plen = 0;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
void rt6_remove_prefsrc(struct inet6_ifaddr *ifp)
|
|
|
|
{
|
|
|
|
struct net *net = dev_net(ifp->idev->dev);
|
|
|
|
struct arg_dev_net_ip adni = {
|
|
|
|
.dev = ifp->idev->dev,
|
|
|
|
.net = net,
|
|
|
|
.addr = &ifp->addr,
|
|
|
|
};
|
|
|
|
fib6_clean_all(net, fib6_remove_prefsrc, 0, &adni);
|
|
|
|
}
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
struct arg_dev_net {
|
|
|
|
struct net_device *dev;
|
|
|
|
struct net *net;
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
static int fib6_ifdown(struct rt6_info *rt, void *arg)
|
|
|
|
{
|
2010-12-16 10:42:40 -07:00
|
|
|
const struct arg_dev_net *adn = arg;
|
|
|
|
const struct net_device *dev = adn->dev;
|
2008-03-04 14:48:30 -07:00
|
|
|
|
2011-12-28 18:19:20 -07:00
|
|
|
if ((rt->dst.dev == dev || !dev) &&
|
2011-12-26 13:24:36 -07:00
|
|
|
rt != adn->net->ipv6.ip6_null_entry)
|
2005-04-16 16:20:36 -06:00
|
|
|
return -1;
|
2011-12-26 13:24:36 -07:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-03-04 00:27:06 -07:00
|
|
|
void rt6_ifdown(struct net *net, struct net_device *dev)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2008-03-04 14:48:30 -07:00
|
|
|
struct arg_dev_net adn = {
|
|
|
|
.dev = dev,
|
|
|
|
.net = net,
|
|
|
|
};
|
|
|
|
|
|
|
|
fib6_clean_all(net, fib6_ifdown, 0, &adn);
|
2008-09-10 18:27:15 -06:00
|
|
|
icmp6_clean_all(fib6_ifdown, &adn);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2012-04-14 23:58:06 -06:00
|
|
|
struct rt6_mtu_change_arg {
|
2005-04-16 16:20:36 -06:00
|
|
|
struct net_device *dev;
|
2012-04-14 23:58:06 -06:00
|
|
|
unsigned int mtu;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg)
|
|
|
|
{
|
|
|
|
struct rt6_mtu_change_arg *arg = (struct rt6_mtu_change_arg *) p_arg;
|
|
|
|
struct inet6_dev *idev;
|
|
|
|
|
|
|
|
/* In IPv6 pmtu discovery is not optional,
|
|
|
|
so that RTAX_MTU lock cannot disable it.
|
|
|
|
We still use this lock to block changes
|
|
|
|
caused by addrconf/ndisc.
|
|
|
|
*/
|
|
|
|
|
|
|
|
idev = __in6_dev_get(arg->dev);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!idev)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
/* For administrative MTU increase, there is no way to discover
|
|
|
|
IPv6 PMTU increase, so PMTU increase should be updated here.
|
|
|
|
Since RFC 1981 doesn't include administrative MTU increase
|
|
|
|
update PMTU increase is a MUST. (i.e. jumbo frame)
|
|
|
|
*/
|
|
|
|
/*
|
|
|
|
If new MTU is less than route PMTU, this new MTU will be the
|
|
|
|
lowest MTU in the path, update the route PMTU to reflect PMTU
|
|
|
|
decreases; if new MTU is greater than route PMTU, and the
|
|
|
|
old MTU is the lowest MTU in the path, update the route PMTU
|
|
|
|
to reflect the increase. In this case if the other nodes' MTU
|
|
|
|
also have the lowest MTU, TOO BIG MESSAGE will be lead to
|
|
|
|
PMTU discouvery.
|
|
|
|
*/
|
2011-12-28 18:19:20 -07:00
|
|
|
if (rt->dst.dev == arg->dev &&
|
2010-06-11 00:31:35 -06:00
|
|
|
!dst_metric_locked(&rt->dst, RTAX_MTU) &&
|
|
|
|
(dst_mtu(&rt->dst) >= arg->mtu ||
|
|
|
|
(dst_mtu(&rt->dst) < arg->mtu &&
|
|
|
|
dst_mtu(&rt->dst) == idev->cnf.mtu6))) {
|
2010-12-08 22:16:57 -07:00
|
|
|
dst_metric_set(&rt->dst, RTAX_MTU, arg->mtu);
|
2007-07-26 01:09:55 -06:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2012-04-14 23:58:06 -06:00
|
|
|
void rt6_mtu_change(struct net_device *dev, unsigned int mtu)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-08-05 00:20:06 -06:00
|
|
|
struct rt6_mtu_change_arg arg = {
|
|
|
|
.dev = dev,
|
|
|
|
.mtu = mtu,
|
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-03-25 06:47:49 -06:00
|
|
|
fib6_clean_all(dev_net(dev), rt6_mtu_change_route, 0, &arg);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2007-06-05 13:38:30 -06:00
|
|
|
static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = {
|
2006-08-26 21:13:18 -06:00
|
|
|
[RTA_GATEWAY] = { .len = sizeof(struct in6_addr) },
|
2006-08-22 01:01:08 -06:00
|
|
|
[RTA_OIF] = { .type = NLA_U32 },
|
2006-08-22 01:01:47 -06:00
|
|
|
[RTA_IIF] = { .type = NLA_U32 },
|
2006-08-22 01:01:08 -06:00
|
|
|
[RTA_PRIORITY] = { .type = NLA_U32 },
|
|
|
|
[RTA_METRICS] = { .type = NLA_NESTED },
|
2012-10-21 21:42:09 -06:00
|
|
|
[RTA_MULTIPATH] = { .len = sizeof(struct rtnexthop) },
|
2006-08-22 01:01:08 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh,
|
|
|
|
struct fib6_config *cfg)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-08-22 01:01:08 -06:00
|
|
|
struct rtmsg *rtm;
|
|
|
|
struct nlattr *tb[RTA_MAX+1];
|
|
|
|
int err;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv6_policy);
|
|
|
|
if (err < 0)
|
|
|
|
goto errout;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
err = -EINVAL;
|
|
|
|
rtm = nlmsg_data(nlh);
|
|
|
|
memset(cfg, 0, sizeof(*cfg));
|
|
|
|
|
|
|
|
cfg->fc_table = rtm->rtm_table;
|
|
|
|
cfg->fc_dst_len = rtm->rtm_dst_len;
|
|
|
|
cfg->fc_src_len = rtm->rtm_src_len;
|
|
|
|
cfg->fc_flags = RTF_UP;
|
|
|
|
cfg->fc_protocol = rtm->rtm_protocol;
|
2012-09-04 20:12:42 -06:00
|
|
|
cfg->fc_type = rtm->rtm_type;
|
2006-08-22 01:01:08 -06:00
|
|
|
|
2012-09-04 20:12:42 -06:00
|
|
|
if (rtm->rtm_type == RTN_UNREACHABLE ||
|
|
|
|
rtm->rtm_type == RTN_BLACKHOLE ||
|
2012-09-05 23:53:35 -06:00
|
|
|
rtm->rtm_type == RTN_PROHIBIT ||
|
|
|
|
rtm->rtm_type == RTN_THROW)
|
2006-08-22 01:01:08 -06:00
|
|
|
cfg->fc_flags |= RTF_REJECT;
|
|
|
|
|
2010-09-26 18:07:02 -06:00
|
|
|
if (rtm->rtm_type == RTN_LOCAL)
|
|
|
|
cfg->fc_flags |= RTF_LOCAL;
|
|
|
|
|
2012-09-07 14:12:54 -06:00
|
|
|
cfg->fc_nlinfo.portid = NETLINK_CB(skb).portid;
|
2006-08-22 01:01:08 -06:00
|
|
|
cfg->fc_nlinfo.nlh = nlh;
|
2008-03-25 11:26:21 -06:00
|
|
|
cfg->fc_nlinfo.nl_net = sock_net(skb->sk);
|
2006-08-22 01:01:08 -06:00
|
|
|
|
|
|
|
if (tb[RTA_GATEWAY]) {
|
|
|
|
nla_memcpy(&cfg->fc_gateway, tb[RTA_GATEWAY], 16);
|
|
|
|
cfg->fc_flags |= RTF_GATEWAY;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2006-08-22 01:01:08 -06:00
|
|
|
|
|
|
|
if (tb[RTA_DST]) {
|
|
|
|
int plen = (rtm->rtm_dst_len + 7) >> 3;
|
|
|
|
|
|
|
|
if (nla_len(tb[RTA_DST]) < plen)
|
|
|
|
goto errout;
|
|
|
|
|
|
|
|
nla_memcpy(&cfg->fc_dst, tb[RTA_DST], plen);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2006-08-22 01:01:08 -06:00
|
|
|
|
|
|
|
if (tb[RTA_SRC]) {
|
|
|
|
int plen = (rtm->rtm_src_len + 7) >> 3;
|
|
|
|
|
|
|
|
if (nla_len(tb[RTA_SRC]) < plen)
|
|
|
|
goto errout;
|
|
|
|
|
|
|
|
nla_memcpy(&cfg->fc_src, tb[RTA_SRC], plen);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2006-08-22 01:01:08 -06:00
|
|
|
|
2011-04-13 15:10:57 -06:00
|
|
|
if (tb[RTA_PREFSRC])
|
|
|
|
nla_memcpy(&cfg->fc_prefsrc, tb[RTA_PREFSRC], 16);
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
if (tb[RTA_OIF])
|
|
|
|
cfg->fc_ifindex = nla_get_u32(tb[RTA_OIF]);
|
|
|
|
|
|
|
|
if (tb[RTA_PRIORITY])
|
|
|
|
cfg->fc_metric = nla_get_u32(tb[RTA_PRIORITY]);
|
|
|
|
|
|
|
|
if (tb[RTA_METRICS]) {
|
|
|
|
cfg->fc_mx = nla_data(tb[RTA_METRICS]);
|
|
|
|
cfg->fc_mx_len = nla_len(tb[RTA_METRICS]);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2006-08-22 01:01:08 -06:00
|
|
|
|
|
|
|
if (tb[RTA_TABLE])
|
|
|
|
cfg->fc_table = nla_get_u32(tb[RTA_TABLE]);
|
|
|
|
|
2012-10-21 21:42:09 -06:00
|
|
|
if (tb[RTA_MULTIPATH]) {
|
|
|
|
cfg->fc_mp = nla_data(tb[RTA_MULTIPATH]);
|
|
|
|
cfg->fc_mp_len = nla_len(tb[RTA_MULTIPATH]);
|
|
|
|
}
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
err = 0;
|
|
|
|
errout:
|
|
|
|
return err;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2012-10-21 21:42:09 -06:00
|
|
|
static int ip6_route_multipath(struct fib6_config *cfg, int add)
|
|
|
|
{
|
|
|
|
struct fib6_config r_cfg;
|
|
|
|
struct rtnexthop *rtnh;
|
|
|
|
int remaining;
|
|
|
|
int attrlen;
|
|
|
|
int err = 0, last_err = 0;
|
|
|
|
|
|
|
|
beginning:
|
|
|
|
rtnh = (struct rtnexthop *)cfg->fc_mp;
|
|
|
|
remaining = cfg->fc_mp_len;
|
|
|
|
|
|
|
|
/* Parse a Multipath Entry */
|
|
|
|
while (rtnh_ok(rtnh, remaining)) {
|
|
|
|
memcpy(&r_cfg, cfg, sizeof(*cfg));
|
|
|
|
if (rtnh->rtnh_ifindex)
|
|
|
|
r_cfg.fc_ifindex = rtnh->rtnh_ifindex;
|
|
|
|
|
|
|
|
attrlen = rtnh_attrlen(rtnh);
|
|
|
|
if (attrlen > 0) {
|
|
|
|
struct nlattr *nla, *attrs = rtnh_attrs(rtnh);
|
|
|
|
|
|
|
|
nla = nla_find(attrs, attrlen, RTA_GATEWAY);
|
|
|
|
if (nla) {
|
|
|
|
nla_memcpy(&r_cfg.fc_gateway, nla, 16);
|
|
|
|
r_cfg.fc_flags |= RTF_GATEWAY;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
err = add ? ip6_route_add(&r_cfg) : ip6_route_del(&r_cfg);
|
|
|
|
if (err) {
|
|
|
|
last_err = err;
|
|
|
|
/* If we are trying to remove a route, do not stop the
|
|
|
|
* loop when ip6_route_del() fails (because next hop is
|
|
|
|
* already gone), we should try to remove all next hops.
|
|
|
|
*/
|
|
|
|
if (add) {
|
|
|
|
/* If add fails, we should try to delete all
|
|
|
|
* next hops that have been already added.
|
|
|
|
*/
|
|
|
|
add = 0;
|
|
|
|
goto beginning;
|
|
|
|
}
|
|
|
|
}
|
2012-11-01 16:58:22 -06:00
|
|
|
/* Because each route is added like a single route we remove
|
|
|
|
* this flag after the first nexthop (if there is a collision,
|
|
|
|
* we have already fail to add the first nexthop:
|
|
|
|
* fib6_add_rt2node() has reject it).
|
|
|
|
*/
|
|
|
|
cfg->fc_nlinfo.nlh->nlmsg_flags &= ~NLM_F_EXCL;
|
2012-10-21 21:42:09 -06:00
|
|
|
rtnh = rtnh_next(rtnh, &remaining);
|
|
|
|
}
|
|
|
|
|
|
|
|
return last_err;
|
|
|
|
}
|
|
|
|
|
2013-03-21 01:45:29 -06:00
|
|
|
static int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-08-22 01:01:08 -06:00
|
|
|
struct fib6_config cfg;
|
|
|
|
int err;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
err = rtm_to_fib6_config(skb, nlh, &cfg);
|
|
|
|
if (err < 0)
|
|
|
|
return err;
|
|
|
|
|
2012-10-21 21:42:09 -06:00
|
|
|
if (cfg.fc_mp)
|
|
|
|
return ip6_route_multipath(&cfg, 0);
|
|
|
|
else
|
|
|
|
return ip6_route_del(&cfg);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2013-03-21 01:45:29 -06:00
|
|
|
static int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-08-22 01:01:08 -06:00
|
|
|
struct fib6_config cfg;
|
|
|
|
int err;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
err = rtm_to_fib6_config(skb, nlh, &cfg);
|
|
|
|
if (err < 0)
|
|
|
|
return err;
|
|
|
|
|
2012-10-21 21:42:09 -06:00
|
|
|
if (cfg.fc_mp)
|
|
|
|
return ip6_route_multipath(&cfg, 1);
|
|
|
|
else
|
|
|
|
return ip6_route_add(&cfg);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-11-10 15:10:15 -07:00
|
|
|
static inline size_t rt6_nlmsg_size(void)
|
|
|
|
{
|
|
|
|
return NLMSG_ALIGN(sizeof(struct rtmsg))
|
|
|
|
+ nla_total_size(16) /* RTA_SRC */
|
|
|
|
+ nla_total_size(16) /* RTA_DST */
|
|
|
|
+ nla_total_size(16) /* RTA_GATEWAY */
|
|
|
|
+ nla_total_size(16) /* RTA_PREFSRC */
|
|
|
|
+ nla_total_size(4) /* RTA_TABLE */
|
|
|
|
+ nla_total_size(4) /* RTA_IIF */
|
|
|
|
+ nla_total_size(4) /* RTA_OIF */
|
|
|
|
+ nla_total_size(4) /* RTA_PRIORITY */
|
2007-01-23 23:09:41 -07:00
|
|
|
+ RTAX_MAX * nla_total_size(4) /* RTA_METRICS */
|
2006-11-10 15:10:15 -07:00
|
|
|
+ nla_total_size(sizeof(struct rta_cacheinfo));
|
|
|
|
}
|
|
|
|
|
2008-08-14 16:33:21 -06:00
|
|
|
static int rt6_fill_node(struct net *net,
|
|
|
|
struct sk_buff *skb, struct rt6_info *rt,
|
2005-06-21 14:51:04 -06:00
|
|
|
struct in6_addr *dst, struct in6_addr *src,
|
2012-09-07 14:12:54 -06:00
|
|
|
int iif, int type, u32 portid, u32 seq,
|
2008-04-02 18:22:53 -06:00
|
|
|
int prefix, int nowait, unsigned int flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct rtmsg *rtm;
|
2006-08-22 01:01:27 -06:00
|
|
|
struct nlmsghdr *nlh;
|
2006-11-27 10:27:07 -07:00
|
|
|
long expires;
|
2006-08-11 00:09:48 -06:00
|
|
|
u32 table;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
if (prefix) { /* user wants prefix routes only */
|
|
|
|
if (!(rt->rt6i_flags & RTF_PREFIX_RT)) {
|
|
|
|
/* success since this is not a prefix route */
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-09-07 14:12:54 -06:00
|
|
|
nlh = nlmsg_put(skb, portid, seq, type, sizeof(*rtm), flags);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!nlh)
|
2007-02-01 00:16:40 -07:00
|
|
|
return -EMSGSIZE;
|
2006-08-22 01:01:27 -06:00
|
|
|
|
|
|
|
rtm = nlmsg_data(nlh);
|
2005-04-16 16:20:36 -06:00
|
|
|
rtm->rtm_family = AF_INET6;
|
|
|
|
rtm->rtm_dst_len = rt->rt6i_dst.plen;
|
|
|
|
rtm->rtm_src_len = rt->rt6i_src.plen;
|
|
|
|
rtm->rtm_tos = 0;
|
2006-08-05 00:20:06 -06:00
|
|
|
if (rt->rt6i_table)
|
2006-08-11 00:09:48 -06:00
|
|
|
table = rt->rt6i_table->tb6_id;
|
2006-08-05 00:20:06 -06:00
|
|
|
else
|
2006-08-11 00:09:48 -06:00
|
|
|
table = RT6_TABLE_UNSPEC;
|
|
|
|
rtm->rtm_table = table;
|
2012-04-01 18:27:33 -06:00
|
|
|
if (nla_put_u32(skb, RTA_TABLE, table))
|
|
|
|
goto nla_put_failure;
|
2012-09-04 20:12:42 -06:00
|
|
|
if (rt->rt6i_flags & RTF_REJECT) {
|
|
|
|
switch (rt->dst.error) {
|
|
|
|
case -EINVAL:
|
|
|
|
rtm->rtm_type = RTN_BLACKHOLE;
|
|
|
|
break;
|
|
|
|
case -EACCES:
|
|
|
|
rtm->rtm_type = RTN_PROHIBIT;
|
|
|
|
break;
|
2012-09-05 23:53:35 -06:00
|
|
|
case -EAGAIN:
|
|
|
|
rtm->rtm_type = RTN_THROW;
|
|
|
|
break;
|
2012-09-04 20:12:42 -06:00
|
|
|
default:
|
|
|
|
rtm->rtm_type = RTN_UNREACHABLE;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2011-12-03 16:02:47 -07:00
|
|
|
else if (rt->rt6i_flags & RTF_LOCAL)
|
2010-09-26 18:07:02 -06:00
|
|
|
rtm->rtm_type = RTN_LOCAL;
|
2011-12-28 18:19:20 -07:00
|
|
|
else if (rt->dst.dev && (rt->dst.dev->flags & IFF_LOOPBACK))
|
2005-04-16 16:20:36 -06:00
|
|
|
rtm->rtm_type = RTN_LOCAL;
|
|
|
|
else
|
|
|
|
rtm->rtm_type = RTN_UNICAST;
|
|
|
|
rtm->rtm_flags = 0;
|
|
|
|
rtm->rtm_scope = RT_SCOPE_UNIVERSE;
|
|
|
|
rtm->rtm_protocol = rt->rt6i_protocol;
|
2011-12-03 16:02:47 -07:00
|
|
|
if (rt->rt6i_flags & RTF_DYNAMIC)
|
2005-04-16 16:20:36 -06:00
|
|
|
rtm->rtm_protocol = RTPROT_REDIRECT;
|
2012-07-09 22:45:50 -06:00
|
|
|
else if (rt->rt6i_flags & RTF_ADDRCONF) {
|
|
|
|
if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ROUTEINFO))
|
|
|
|
rtm->rtm_protocol = RTPROT_RA;
|
|
|
|
else
|
|
|
|
rtm->rtm_protocol = RTPROT_KERNEL;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-12-03 16:02:47 -07:00
|
|
|
if (rt->rt6i_flags & RTF_CACHE)
|
2005-04-16 16:20:36 -06:00
|
|
|
rtm->rtm_flags |= RTM_F_CLONED;
|
|
|
|
|
|
|
|
if (dst) {
|
2012-04-01 18:27:33 -06:00
|
|
|
if (nla_put(skb, RTA_DST, 16, dst))
|
|
|
|
goto nla_put_failure;
|
2007-02-09 07:24:49 -07:00
|
|
|
rtm->rtm_dst_len = 128;
|
2005-04-16 16:20:36 -06:00
|
|
|
} else if (rtm->rtm_dst_len)
|
2012-04-01 18:27:33 -06:00
|
|
|
if (nla_put(skb, RTA_DST, 16, &rt->rt6i_dst.addr))
|
|
|
|
goto nla_put_failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
#ifdef CONFIG_IPV6_SUBTREES
|
|
|
|
if (src) {
|
2012-04-01 18:27:33 -06:00
|
|
|
if (nla_put(skb, RTA_SRC, 16, src))
|
|
|
|
goto nla_put_failure;
|
2007-02-09 07:24:49 -07:00
|
|
|
rtm->rtm_src_len = 128;
|
2012-04-01 18:27:33 -06:00
|
|
|
} else if (rtm->rtm_src_len &&
|
|
|
|
nla_put(skb, RTA_SRC, 16, &rt->rt6i_src.addr))
|
|
|
|
goto nla_put_failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
#endif
|
2008-04-02 18:22:53 -06:00
|
|
|
if (iif) {
|
|
|
|
#ifdef CONFIG_IPV6_MROUTE
|
|
|
|
if (ipv6_addr_is_multicast(&rt->rt6i_dst.addr)) {
|
2008-12-10 17:30:15 -07:00
|
|
|
int err = ip6mr_get_route(net, skb, rtm, nowait);
|
2008-04-02 18:22:53 -06:00
|
|
|
if (err <= 0) {
|
|
|
|
if (!nowait) {
|
|
|
|
if (err == 0)
|
|
|
|
return 0;
|
|
|
|
goto nla_put_failure;
|
|
|
|
} else {
|
|
|
|
if (err == -EMSGSIZE)
|
|
|
|
goto nla_put_failure;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
} else
|
|
|
|
#endif
|
2012-04-01 18:27:33 -06:00
|
|
|
if (nla_put_u32(skb, RTA_IIF, iif))
|
|
|
|
goto nla_put_failure;
|
2008-04-02 18:22:53 -06:00
|
|
|
} else if (dst) {
|
2005-04-16 16:20:36 -06:00
|
|
|
struct in6_addr saddr_buf;
|
2012-04-01 18:27:33 -06:00
|
|
|
if (ip6_route_get_saddr(net, rt, dst, 0, &saddr_buf) == 0 &&
|
|
|
|
nla_put(skb, RTA_PREFSRC, 16, &saddr_buf))
|
|
|
|
goto nla_put_failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2006-08-22 01:01:27 -06:00
|
|
|
|
2011-04-13 15:10:57 -06:00
|
|
|
if (rt->rt6i_prefsrc.plen) {
|
|
|
|
struct in6_addr saddr_buf;
|
2011-11-20 20:39:03 -07:00
|
|
|
saddr_buf = rt->rt6i_prefsrc.addr;
|
2012-04-01 18:27:33 -06:00
|
|
|
if (nla_put(skb, RTA_PREFSRC, 16, &saddr_buf))
|
|
|
|
goto nla_put_failure;
|
2011-04-13 15:10:57 -06:00
|
|
|
}
|
|
|
|
|
2010-12-08 22:16:57 -07:00
|
|
|
if (rtnetlink_put_metrics(skb, dst_metrics_ptr(&rt->dst)) < 0)
|
2006-08-22 01:01:27 -06:00
|
|
|
goto nla_put_failure;
|
|
|
|
|
2013-01-17 05:53:15 -07:00
|
|
|
if (rt->rt6i_flags & RTF_GATEWAY) {
|
|
|
|
if (nla_put(skb, RTA_GATEWAY, 16, &rt->rt6i_gateway) < 0)
|
2012-03-27 03:53:52 -06:00
|
|
|
goto nla_put_failure;
|
|
|
|
}
|
2006-08-22 01:01:27 -06:00
|
|
|
|
2012-04-01 18:27:33 -06:00
|
|
|
if (rt->dst.dev &&
|
|
|
|
nla_put_u32(skb, RTA_OIF, rt->dst.dev->ifindex))
|
|
|
|
goto nla_put_failure;
|
|
|
|
if (nla_put_u32(skb, RTA_PRIORITY, rt->rt6i_metric))
|
|
|
|
goto nla_put_failure;
|
2012-07-29 10:01:30 -06:00
|
|
|
|
|
|
|
expires = (rt->rt6i_flags & RTF_EXPIRES) ? rt->dst.expires - jiffies : 0;
|
2008-05-19 17:55:13 -06:00
|
|
|
|
2012-07-10 06:06:14 -06:00
|
|
|
if (rtnl_put_cacheinfo(skb, &rt->dst, 0, expires, rt->dst.error) < 0)
|
2006-11-27 10:27:07 -07:00
|
|
|
goto nla_put_failure;
|
2006-08-22 01:01:27 -06:00
|
|
|
|
|
|
|
return nlmsg_end(skb, nlh);
|
|
|
|
|
|
|
|
nla_put_failure:
|
2007-02-01 00:16:40 -07:00
|
|
|
nlmsg_cancel(skb, nlh);
|
|
|
|
return -EMSGSIZE;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-08-11 00:11:17 -06:00
|
|
|
int rt6_dump_route(struct rt6_info *rt, void *p_arg)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct rt6_rtnl_dump_arg *arg = (struct rt6_rtnl_dump_arg *) p_arg;
|
|
|
|
int prefix;
|
|
|
|
|
2006-08-22 01:01:27 -06:00
|
|
|
if (nlmsg_len(arg->cb->nlh) >= sizeof(struct rtmsg)) {
|
|
|
|
struct rtmsg *rtm = nlmsg_data(arg->cb->nlh);
|
2005-04-16 16:20:36 -06:00
|
|
|
prefix = (rtm->rtm_flags & RTM_F_PREFIX) != 0;
|
|
|
|
} else
|
|
|
|
prefix = 0;
|
|
|
|
|
2008-08-14 16:33:21 -06:00
|
|
|
return rt6_fill_node(arg->net,
|
|
|
|
arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE,
|
2012-09-07 14:12:54 -06:00
|
|
|
NETLINK_CB(arg->cb->skb).portid, arg->cb->nlh->nlmsg_seq,
|
2008-04-02 18:22:53 -06:00
|
|
|
prefix, 0, NLM_F_MULTI);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2013-03-21 01:45:29 -06:00
|
|
|
static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2008-03-25 11:26:21 -06:00
|
|
|
struct net *net = sock_net(in_skb->sk);
|
2006-08-22 01:01:47 -06:00
|
|
|
struct nlattr *tb[RTA_MAX+1];
|
|
|
|
struct rt6_info *rt;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct sk_buff *skb;
|
2006-08-22 01:01:47 -06:00
|
|
|
struct rtmsg *rtm;
|
2011-03-12 14:22:43 -07:00
|
|
|
struct flowi6 fl6;
|
2012-03-31 22:03:45 -06:00
|
|
|
int err, iif = 0, oif = 0;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:47 -06:00
|
|
|
err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv6_policy);
|
|
|
|
if (err < 0)
|
|
|
|
goto errout;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:47 -06:00
|
|
|
err = -EINVAL;
|
2011-03-12 14:22:43 -07:00
|
|
|
memset(&fl6, 0, sizeof(fl6));
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:47 -06:00
|
|
|
if (tb[RTA_SRC]) {
|
|
|
|
if (nla_len(tb[RTA_SRC]) < sizeof(struct in6_addr))
|
|
|
|
goto errout;
|
|
|
|
|
2011-11-20 20:39:03 -07:00
|
|
|
fl6.saddr = *(struct in6_addr *)nla_data(tb[RTA_SRC]);
|
2006-08-22 01:01:47 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
if (tb[RTA_DST]) {
|
|
|
|
if (nla_len(tb[RTA_DST]) < sizeof(struct in6_addr))
|
|
|
|
goto errout;
|
|
|
|
|
2011-11-20 20:39:03 -07:00
|
|
|
fl6.daddr = *(struct in6_addr *)nla_data(tb[RTA_DST]);
|
2006-08-22 01:01:47 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
if (tb[RTA_IIF])
|
|
|
|
iif = nla_get_u32(tb[RTA_IIF]);
|
|
|
|
|
|
|
|
if (tb[RTA_OIF])
|
2012-03-31 22:03:45 -06:00
|
|
|
oif = nla_get_u32(tb[RTA_OIF]);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
if (iif) {
|
|
|
|
struct net_device *dev;
|
2012-03-31 22:03:45 -06:00
|
|
|
int flags = 0;
|
|
|
|
|
2008-03-04 14:47:47 -07:00
|
|
|
dev = __dev_get_by_index(net, iif);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (!dev) {
|
|
|
|
err = -ENODEV;
|
2006-08-22 01:01:47 -06:00
|
|
|
goto errout;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2012-03-31 22:03:45 -06:00
|
|
|
|
|
|
|
fl6.flowi6_iif = iif;
|
|
|
|
|
|
|
|
if (!ipv6_addr_any(&fl6.saddr))
|
|
|
|
flags |= RT6_LOOKUP_F_HAS_SADDR;
|
|
|
|
|
|
|
|
rt = (struct rt6_info *)ip6_route_input_lookup(net, dev, &fl6,
|
|
|
|
flags);
|
|
|
|
} else {
|
|
|
|
fl6.flowi6_oif = oif;
|
|
|
|
|
|
|
|
rt = (struct rt6_info *)ip6_route_output(net, NULL, &fl6);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-08-22 01:01:47 -06:00
|
|
|
skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!skb) {
|
2012-10-28 18:13:19 -06:00
|
|
|
ip6_rt_put(rt);
|
2006-08-22 01:01:47 -06:00
|
|
|
err = -ENOBUFS;
|
|
|
|
goto errout;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-22 01:01:47 -06:00
|
|
|
/* Reserve room for dummy headers, this skb can pass
|
|
|
|
through good chunk of routing engine.
|
|
|
|
*/
|
2007-03-19 16:30:44 -06:00
|
|
|
skb_reset_mac_header(skb);
|
2006-08-22 01:01:47 -06:00
|
|
|
skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr));
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-06-11 00:31:35 -06:00
|
|
|
skb_dst_set(skb, &rt->dst);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-03-12 14:22:43 -07:00
|
|
|
err = rt6_fill_node(net, skb, rt, &fl6.daddr, &fl6.saddr, iif,
|
2012-09-07 14:12:54 -06:00
|
|
|
RTM_NEWROUTE, NETLINK_CB(in_skb).portid,
|
2008-04-02 18:22:53 -06:00
|
|
|
nlh->nlmsg_seq, 0, 0, 0);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (err < 0) {
|
2006-08-22 01:01:47 -06:00
|
|
|
kfree_skb(skb);
|
|
|
|
goto errout;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2012-09-07 14:12:54 -06:00
|
|
|
err = rtnl_unicast(skb, net, NETLINK_CB(in_skb).portid);
|
2006-08-22 01:01:47 -06:00
|
|
|
errout:
|
2005-04-16 16:20:36 -06:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2006-08-22 01:01:08 -06:00
|
|
|
void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct sk_buff *skb;
|
2008-03-04 14:47:47 -07:00
|
|
|
struct net *net = info->nl_net;
|
2007-12-13 10:45:12 -07:00
|
|
|
u32 seq;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
err = -ENOBUFS;
|
2011-12-03 16:02:47 -07:00
|
|
|
seq = info->nlh ? info->nlh->nlmsg_seq : 0;
|
2006-08-22 01:01:08 -06:00
|
|
|
|
2006-11-10 15:10:15 -07:00
|
|
|
skb = nlmsg_new(rt6_nlmsg_size(), gfp_any());
|
2011-12-03 16:02:47 -07:00
|
|
|
if (!skb)
|
2006-08-15 01:35:24 -06:00
|
|
|
goto errout;
|
|
|
|
|
2008-08-14 16:33:21 -06:00
|
|
|
err = rt6_fill_node(net, skb, rt, NULL, NULL, 0,
|
2012-09-07 14:12:54 -06:00
|
|
|
event, info->portid, seq, 0, 0, 0);
|
2007-02-01 00:16:40 -07:00
|
|
|
if (err < 0) {
|
|
|
|
/* -EMSGSIZE implies BUG in rt6_nlmsg_size() */
|
|
|
|
WARN_ON(err == -EMSGSIZE);
|
|
|
|
kfree_skb(skb);
|
|
|
|
goto errout;
|
|
|
|
}
|
2012-09-07 14:12:54 -06:00
|
|
|
rtnl_notify(skb, net, info->portid, RTNLGRP_IPV6_ROUTE,
|
2009-02-25 00:18:28 -07:00
|
|
|
info->nlh, gfp_any());
|
|
|
|
return;
|
2006-08-15 01:35:24 -06:00
|
|
|
errout:
|
|
|
|
if (err < 0)
|
2008-03-04 14:47:47 -07:00
|
|
|
rtnl_set_sk_err(net, RTNLGRP_IPV6_ROUTE, err);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
static int ip6_route_dev_notify(struct notifier_block *this,
|
2013-05-27 19:30:21 -06:00
|
|
|
unsigned long event, void *ptr)
|
2008-03-04 14:48:30 -07:00
|
|
|
{
|
2013-05-27 19:30:21 -06:00
|
|
|
struct net_device *dev = netdev_notifier_info_to_dev(ptr);
|
2008-03-25 06:47:49 -06:00
|
|
|
struct net *net = dev_net(dev);
|
2008-03-04 14:48:30 -07:00
|
|
|
|
|
|
|
if (event == NETDEV_REGISTER && (dev->flags & IFF_LOOPBACK)) {
|
2010-06-11 00:31:35 -06:00
|
|
|
net->ipv6.ip6_null_entry->dst.dev = dev;
|
2008-03-04 14:48:30 -07:00
|
|
|
net->ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(dev);
|
|
|
|
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
|
2010-06-11 00:31:35 -06:00
|
|
|
net->ipv6.ip6_prohibit_entry->dst.dev = dev;
|
2008-03-04 14:48:30 -07:00
|
|
|
net->ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(dev);
|
2010-06-11 00:31:35 -06:00
|
|
|
net->ipv6.ip6_blk_hole_entry->dst.dev = dev;
|
2008-03-04 14:48:30 -07:00
|
|
|
net->ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(dev);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
|
|
|
return NOTIFY_OK;
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* /proc
|
|
|
|
*/
|
|
|
|
|
|
|
|
#ifdef CONFIG_PROC_FS
|
|
|
|
|
2007-11-06 06:27:11 -07:00
|
|
|
static const struct file_operations ipv6_route_proc_fops = {
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.open = ipv6_route_open,
|
|
|
|
.read = seq_read,
|
|
|
|
.llseek = seq_lseek,
|
2013-09-21 08:55:59 -06:00
|
|
|
.release = seq_release_net,
|
2007-11-06 06:27:11 -07:00
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
static int rt6_stats_seq_show(struct seq_file *seq, void *v)
|
|
|
|
{
|
2008-03-04 14:46:23 -07:00
|
|
|
struct net *net = (struct net *)seq->private;
|
2005-04-16 16:20:36 -06:00
|
|
|
seq_printf(seq, "%04x %04x %04x %04x %04x %04x %04x\n",
|
2008-03-04 14:46:23 -07:00
|
|
|
net->ipv6.rt6_stats->fib_nodes,
|
|
|
|
net->ipv6.rt6_stats->fib_route_nodes,
|
|
|
|
net->ipv6.rt6_stats->fib_rt_alloc,
|
|
|
|
net->ipv6.rt6_stats->fib_rt_entries,
|
|
|
|
net->ipv6.rt6_stats->fib_rt_cache,
|
2010-10-08 00:37:34 -06:00
|
|
|
dst_entries_get_slow(&net->ipv6.ip6_dst_ops),
|
2008-03-04 14:46:23 -07:00
|
|
|
net->ipv6.rt6_stats->fib_discarded_routes);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int rt6_stats_seq_open(struct inode *inode, struct file *file)
|
|
|
|
{
|
2008-07-18 05:07:21 -06:00
|
|
|
return single_open_net(inode, file, rt6_stats_seq_show);
|
2008-03-04 14:46:23 -07:00
|
|
|
}
|
|
|
|
|
2007-02-12 01:55:35 -07:00
|
|
|
static const struct file_operations rt6_stats_seq_fops = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.open = rt6_stats_seq_open,
|
|
|
|
.read = seq_read,
|
|
|
|
.llseek = seq_lseek,
|
2008-07-18 05:07:44 -06:00
|
|
|
.release = single_release_net,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
#endif /* CONFIG_PROC_FS */
|
|
|
|
|
|
|
|
#ifdef CONFIG_SYSCTL
|
|
|
|
|
|
|
|
static
|
2013-06-12 00:04:25 -06:00
|
|
|
int ipv6_sysctl_rtcache_flush(struct ctl_table *ctl, int write,
|
2005-04-16 16:20:36 -06:00
|
|
|
void __user *buffer, size_t *lenp, loff_t *ppos)
|
|
|
|
{
|
2011-02-24 12:48:03 -07:00
|
|
|
struct net *net;
|
|
|
|
int delay;
|
|
|
|
if (!write)
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EINVAL;
|
2011-02-24 12:48:03 -07:00
|
|
|
|
|
|
|
net = (struct net *)ctl->extra1;
|
|
|
|
delay = net->ipv6.sysctl.flush_delay;
|
|
|
|
proc_dointvec(ctl, write, buffer, lenp, ppos);
|
2013-08-01 02:04:14 -06:00
|
|
|
fib6_run_gc(delay <= 0 ? 0 : (unsigned long)delay, net, delay > 0);
|
2011-02-24 12:48:03 -07:00
|
|
|
return 0;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2013-06-12 00:04:25 -06:00
|
|
|
struct ctl_table ipv6_route_table_template[] = {
|
2007-02-09 07:24:49 -07:00
|
|
|
{
|
2005-04-16 16:20:36 -06:00
|
|
|
.procname = "flush",
|
2008-01-10 04:01:01 -07:00
|
|
|
.data = &init_net.ipv6.sysctl.flush_delay,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
2005-04-28 13:11:49 -06:00
|
|
|
.mode = 0200,
|
2008-11-03 19:21:05 -07:00
|
|
|
.proc_handler = ipv6_sysctl_rtcache_flush
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.procname = "gc_thresh",
|
2008-03-04 14:48:53 -07:00
|
|
|
.data = &ip6_dst_ops_template.gc_thresh,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
|
|
|
.mode = 0644,
|
2008-11-03 19:21:05 -07:00
|
|
|
.proc_handler = proc_dointvec,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.procname = "max_size",
|
2008-01-10 04:01:01 -07:00
|
|
|
.data = &init_net.ipv6.sysctl.ip6_rt_max_size,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
|
|
|
.mode = 0644,
|
2008-11-03 19:21:05 -07:00
|
|
|
.proc_handler = proc_dointvec,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.procname = "gc_min_interval",
|
2008-01-10 04:01:01 -07:00
|
|
|
.data = &init_net.ipv6.sysctl.ip6_rt_gc_min_interval,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
|
|
|
.mode = 0644,
|
2008-11-03 19:21:05 -07:00
|
|
|
.proc_handler = proc_dointvec_jiffies,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.procname = "gc_timeout",
|
2008-01-10 04:01:01 -07:00
|
|
|
.data = &init_net.ipv6.sysctl.ip6_rt_gc_timeout,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
|
|
|
.mode = 0644,
|
2008-11-03 19:21:05 -07:00
|
|
|
.proc_handler = proc_dointvec_jiffies,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.procname = "gc_interval",
|
2008-01-10 04:01:01 -07:00
|
|
|
.data = &init_net.ipv6.sysctl.ip6_rt_gc_interval,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
|
|
|
.mode = 0644,
|
2008-11-03 19:21:05 -07:00
|
|
|
.proc_handler = proc_dointvec_jiffies,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.procname = "gc_elasticity",
|
2008-01-10 04:01:01 -07:00
|
|
|
.data = &init_net.ipv6.sysctl.ip6_rt_gc_elasticity,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
|
|
|
.mode = 0644,
|
2010-08-14 23:42:51 -06:00
|
|
|
.proc_handler = proc_dointvec,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.procname = "mtu_expires",
|
2008-01-10 04:01:01 -07:00
|
|
|
.data = &init_net.ipv6.sysctl.ip6_rt_mtu_expires,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
|
|
|
.mode = 0644,
|
2008-11-03 19:21:05 -07:00
|
|
|
.proc_handler = proc_dointvec_jiffies,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.procname = "min_adv_mss",
|
2008-01-10 04:01:01 -07:00
|
|
|
.data = &init_net.ipv6.sysctl.ip6_rt_min_advmss,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
|
|
|
.mode = 0644,
|
2010-08-14 23:42:51 -06:00
|
|
|
.proc_handler = proc_dointvec,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.procname = "gc_min_interval_ms",
|
2008-01-10 04:01:01 -07:00
|
|
|
.data = &init_net.ipv6.sysctl.ip6_rt_gc_min_interval,
|
2005-04-16 16:20:36 -06:00
|
|
|
.maxlen = sizeof(int),
|
|
|
|
.mode = 0644,
|
2008-11-03 19:21:05 -07:00
|
|
|
.proc_handler = proc_dointvec_ms_jiffies,
|
2005-04-16 16:20:36 -06:00
|
|
|
},
|
2009-11-05 14:32:03 -07:00
|
|
|
{ }
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2010-01-16 20:35:32 -07:00
|
|
|
struct ctl_table * __net_init ipv6_route_sysctl_init(struct net *net)
|
2008-01-10 03:53:43 -07:00
|
|
|
{
|
|
|
|
struct ctl_table *table;
|
|
|
|
|
|
|
|
table = kmemdup(ipv6_route_table_template,
|
|
|
|
sizeof(ipv6_route_table_template),
|
|
|
|
GFP_KERNEL);
|
2008-02-27 08:24:28 -07:00
|
|
|
|
|
|
|
if (table) {
|
|
|
|
table[0].data = &net->ipv6.sysctl.flush_delay;
|
2011-02-24 12:48:03 -07:00
|
|
|
table[0].extra1 = net;
|
2009-08-28 19:34:49 -06:00
|
|
|
table[1].data = &net->ipv6.ip6_dst_ops.gc_thresh;
|
2008-02-27 08:24:28 -07:00
|
|
|
table[2].data = &net->ipv6.sysctl.ip6_rt_max_size;
|
|
|
|
table[3].data = &net->ipv6.sysctl.ip6_rt_gc_min_interval;
|
|
|
|
table[4].data = &net->ipv6.sysctl.ip6_rt_gc_timeout;
|
|
|
|
table[5].data = &net->ipv6.sysctl.ip6_rt_gc_interval;
|
|
|
|
table[6].data = &net->ipv6.sysctl.ip6_rt_gc_elasticity;
|
|
|
|
table[7].data = &net->ipv6.sysctl.ip6_rt_mtu_expires;
|
|
|
|
table[8].data = &net->ipv6.sysctl.ip6_rt_min_advmss;
|
2009-12-18 21:11:03 -07:00
|
|
|
table[9].data = &net->ipv6.sysctl.ip6_rt_gc_min_interval;
|
2012-11-15 20:02:59 -07:00
|
|
|
|
|
|
|
/* Don't export sysctls to unprivileged users */
|
|
|
|
if (net->user_ns != &init_user_ns)
|
|
|
|
table[0].procname = NULL;
|
2008-02-27 08:24:28 -07:00
|
|
|
}
|
|
|
|
|
2008-01-10 03:53:43 -07:00
|
|
|
return table;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
#endif
|
|
|
|
|
2010-01-16 20:35:32 -07:00
|
|
|
static int __net_init ip6_route_net_init(struct net *net)
|
2008-03-04 14:45:33 -07:00
|
|
|
{
|
2008-04-21 15:25:23 -06:00
|
|
|
int ret = -ENOMEM;
|
2008-03-04 14:48:30 -07:00
|
|
|
|
2009-08-28 19:34:49 -06:00
|
|
|
memcpy(&net->ipv6.ip6_dst_ops, &ip6_dst_ops_template,
|
|
|
|
sizeof(net->ipv6.ip6_dst_ops));
|
2008-03-04 14:49:23 -07:00
|
|
|
|
2010-10-08 00:37:34 -06:00
|
|
|
if (dst_entries_init(&net->ipv6.ip6_dst_ops) < 0)
|
|
|
|
goto out_ip6_dst_ops;
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
net->ipv6.ip6_null_entry = kmemdup(&ip6_null_entry_template,
|
|
|
|
sizeof(*net->ipv6.ip6_null_entry),
|
|
|
|
GFP_KERNEL);
|
|
|
|
if (!net->ipv6.ip6_null_entry)
|
2010-10-08 00:37:34 -06:00
|
|
|
goto out_ip6_dst_entries;
|
2010-06-11 00:31:35 -06:00
|
|
|
net->ipv6.ip6_null_entry->dst.path =
|
2008-03-04 14:48:30 -07:00
|
|
|
(struct dst_entry *)net->ipv6.ip6_null_entry;
|
2010-06-11 00:31:35 -06:00
|
|
|
net->ipv6.ip6_null_entry->dst.ops = &net->ipv6.ip6_dst_ops;
|
net: Implement read-only protection and COW'ing of metrics.
Routing metrics are now copy-on-write.
Initially a route entry points it's metrics at a read-only location.
If a routing table entry exists, it will point there. Else it will
point at the all zero metric place-holder called 'dst_default_metrics'.
The writeability state of the metrics is stored in the low bits of the
metrics pointer, we have two bits left to spare if we want to store
more states.
For the initial implementation, COW is implemented simply via kmalloc.
However future enhancements will change this to place the writable
metrics somewhere else, in order to increase sharing. Very likely
this "somewhere else" will be the inetpeer cache.
Note also that this means that metrics updates may transiently fail
if we cannot COW the metrics successfully.
But even by itself, this patch should decrease memory usage and
increase cache locality especially for routing workloads. In those
cases the read-only metric copies stay in place and never get written
to.
TCP workloads where metrics get updated, and those rare cases where
PMTU triggers occur, will take a very slight performance hit. But
that hit will be alleviated when the long-term writable metrics
move to a more sharable location.
Since the metrics storage went from a u32 array of RTAX_MAX entries to
what is essentially a pointer, some retooling of the dst_entry layout
was necessary.
Most importantly, we need to preserve the alignment of the reference
count so that it doesn't share cache lines with the read-mostly state,
as per Eric Dumazet's alignment assertion checks.
The only non-trivial bit here is the move of the 'flags' member into
the writeable cacheline. This is OK since we are always accessing the
flags around the same moment when we made a modification to the
reference count.
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-26 21:51:05 -07:00
|
|
|
dst_init_metrics(&net->ipv6.ip6_null_entry->dst,
|
|
|
|
ip6_template_metrics, true);
|
2008-03-04 14:48:30 -07:00
|
|
|
|
|
|
|
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
|
|
|
|
net->ipv6.ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template,
|
|
|
|
sizeof(*net->ipv6.ip6_prohibit_entry),
|
|
|
|
GFP_KERNEL);
|
2008-10-07 15:12:10 -06:00
|
|
|
if (!net->ipv6.ip6_prohibit_entry)
|
|
|
|
goto out_ip6_null_entry;
|
2010-06-11 00:31:35 -06:00
|
|
|
net->ipv6.ip6_prohibit_entry->dst.path =
|
2008-03-04 14:48:30 -07:00
|
|
|
(struct dst_entry *)net->ipv6.ip6_prohibit_entry;
|
2010-06-11 00:31:35 -06:00
|
|
|
net->ipv6.ip6_prohibit_entry->dst.ops = &net->ipv6.ip6_dst_ops;
|
net: Implement read-only protection and COW'ing of metrics.
Routing metrics are now copy-on-write.
Initially a route entry points it's metrics at a read-only location.
If a routing table entry exists, it will point there. Else it will
point at the all zero metric place-holder called 'dst_default_metrics'.
The writeability state of the metrics is stored in the low bits of the
metrics pointer, we have two bits left to spare if we want to store
more states.
For the initial implementation, COW is implemented simply via kmalloc.
However future enhancements will change this to place the writable
metrics somewhere else, in order to increase sharing. Very likely
this "somewhere else" will be the inetpeer cache.
Note also that this means that metrics updates may transiently fail
if we cannot COW the metrics successfully.
But even by itself, this patch should decrease memory usage and
increase cache locality especially for routing workloads. In those
cases the read-only metric copies stay in place and never get written
to.
TCP workloads where metrics get updated, and those rare cases where
PMTU triggers occur, will take a very slight performance hit. But
that hit will be alleviated when the long-term writable metrics
move to a more sharable location.
Since the metrics storage went from a u32 array of RTAX_MAX entries to
what is essentially a pointer, some retooling of the dst_entry layout
was necessary.
Most importantly, we need to preserve the alignment of the reference
count so that it doesn't share cache lines with the read-mostly state,
as per Eric Dumazet's alignment assertion checks.
The only non-trivial bit here is the move of the 'flags' member into
the writeable cacheline. This is OK since we are always accessing the
flags around the same moment when we made a modification to the
reference count.
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-26 21:51:05 -07:00
|
|
|
dst_init_metrics(&net->ipv6.ip6_prohibit_entry->dst,
|
|
|
|
ip6_template_metrics, true);
|
2008-03-04 14:48:30 -07:00
|
|
|
|
|
|
|
net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template,
|
|
|
|
sizeof(*net->ipv6.ip6_blk_hole_entry),
|
|
|
|
GFP_KERNEL);
|
2008-10-07 15:12:10 -06:00
|
|
|
if (!net->ipv6.ip6_blk_hole_entry)
|
|
|
|
goto out_ip6_prohibit_entry;
|
2010-06-11 00:31:35 -06:00
|
|
|
net->ipv6.ip6_blk_hole_entry->dst.path =
|
2008-03-04 14:48:30 -07:00
|
|
|
(struct dst_entry *)net->ipv6.ip6_blk_hole_entry;
|
2010-06-11 00:31:35 -06:00
|
|
|
net->ipv6.ip6_blk_hole_entry->dst.ops = &net->ipv6.ip6_dst_ops;
|
net: Implement read-only protection and COW'ing of metrics.
Routing metrics are now copy-on-write.
Initially a route entry points it's metrics at a read-only location.
If a routing table entry exists, it will point there. Else it will
point at the all zero metric place-holder called 'dst_default_metrics'.
The writeability state of the metrics is stored in the low bits of the
metrics pointer, we have two bits left to spare if we want to store
more states.
For the initial implementation, COW is implemented simply via kmalloc.
However future enhancements will change this to place the writable
metrics somewhere else, in order to increase sharing. Very likely
this "somewhere else" will be the inetpeer cache.
Note also that this means that metrics updates may transiently fail
if we cannot COW the metrics successfully.
But even by itself, this patch should decrease memory usage and
increase cache locality especially for routing workloads. In those
cases the read-only metric copies stay in place and never get written
to.
TCP workloads where metrics get updated, and those rare cases where
PMTU triggers occur, will take a very slight performance hit. But
that hit will be alleviated when the long-term writable metrics
move to a more sharable location.
Since the metrics storage went from a u32 array of RTAX_MAX entries to
what is essentially a pointer, some retooling of the dst_entry layout
was necessary.
Most importantly, we need to preserve the alignment of the reference
count so that it doesn't share cache lines with the read-mostly state,
as per Eric Dumazet's alignment assertion checks.
The only non-trivial bit here is the move of the 'flags' member into
the writeable cacheline. This is OK since we are always accessing the
flags around the same moment when we made a modification to the
reference count.
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-26 21:51:05 -07:00
|
|
|
dst_init_metrics(&net->ipv6.ip6_blk_hole_entry->dst,
|
|
|
|
ip6_template_metrics, true);
|
2008-03-04 14:48:30 -07:00
|
|
|
#endif
|
|
|
|
|
2008-10-07 15:15:00 -06:00
|
|
|
net->ipv6.sysctl.flush_delay = 0;
|
|
|
|
net->ipv6.sysctl.ip6_rt_max_size = 4096;
|
|
|
|
net->ipv6.sysctl.ip6_rt_gc_min_interval = HZ / 2;
|
|
|
|
net->ipv6.sysctl.ip6_rt_gc_timeout = 60*HZ;
|
|
|
|
net->ipv6.sysctl.ip6_rt_gc_interval = 30*HZ;
|
|
|
|
net->ipv6.sysctl.ip6_rt_gc_elasticity = 9;
|
|
|
|
net->ipv6.sysctl.ip6_rt_mtu_expires = 10*60*HZ;
|
|
|
|
net->ipv6.sysctl.ip6_rt_min_advmss = IPV6_MIN_MTU - 20 - 40;
|
|
|
|
|
2008-03-04 14:49:47 -07:00
|
|
|
net->ipv6.ip6_rt_gc_expire = 30*HZ;
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
ret = 0;
|
|
|
|
out:
|
|
|
|
return ret;
|
2008-03-04 14:49:23 -07:00
|
|
|
|
2008-10-07 15:12:10 -06:00
|
|
|
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
|
|
|
|
out_ip6_prohibit_entry:
|
|
|
|
kfree(net->ipv6.ip6_prohibit_entry);
|
|
|
|
out_ip6_null_entry:
|
|
|
|
kfree(net->ipv6.ip6_null_entry);
|
|
|
|
#endif
|
2010-10-08 00:37:34 -06:00
|
|
|
out_ip6_dst_entries:
|
|
|
|
dst_entries_destroy(&net->ipv6.ip6_dst_ops);
|
2008-03-04 14:49:23 -07:00
|
|
|
out_ip6_dst_ops:
|
|
|
|
goto out;
|
2008-03-04 14:45:33 -07:00
|
|
|
}
|
|
|
|
|
2010-01-16 20:35:32 -07:00
|
|
|
static void __net_exit ip6_route_net_exit(struct net *net)
|
2008-03-04 14:45:33 -07:00
|
|
|
{
|
2008-03-04 14:48:30 -07:00
|
|
|
kfree(net->ipv6.ip6_null_entry);
|
|
|
|
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
|
|
|
|
kfree(net->ipv6.ip6_prohibit_entry);
|
|
|
|
kfree(net->ipv6.ip6_blk_hole_entry);
|
|
|
|
#endif
|
net dst: fix percpu_counter list corruption and poison overwritten
There're some percpu_counter list corruption and poison overwritten warnings
in recent kernel, which is resulted by fc66f95c.
commit fc66f95c switches to use percpu_counter, in ip6_route_net_init, kernel
init the percpu_counter for dst entries, but, the percpu_counter is never destroyed
in ip6_route_net_exit. So if the related data is freed by kernel, the freed percpu_counter
is still on the list, then if we insert/remove other percpu_counter, list corruption
resulted. Also, if the insert/remove option modifies the ->prev,->next pointer of
the freed value, the poison overwritten is resulted then.
With the following patch, the percpu_counter list corruption and poison overwritten
warnings disappeared.
Signed-off-by: Xiaotian Feng <dfeng@redhat.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: "Pekka Savola (ipv6)" <pekkas@netcore.fi>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-11-02 10:11:05 -06:00
|
|
|
dst_entries_destroy(&net->ipv6.ip6_dst_ops);
|
2008-03-04 14:45:33 -07:00
|
|
|
}
|
|
|
|
|
2012-06-18 06:08:33 -06:00
|
|
|
static int __net_init ip6_route_net_init_late(struct net *net)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_PROC_FS
|
2013-02-17 18:34:54 -07:00
|
|
|
proc_create("ipv6_route", 0, net->proc_net, &ipv6_route_proc_fops);
|
|
|
|
proc_create("rt6_stats", S_IRUGO, net->proc_net, &rt6_stats_seq_fops);
|
2012-06-18 06:08:33 -06:00
|
|
|
#endif
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __net_exit ip6_route_net_exit_late(struct net *net)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_PROC_FS
|
2013-02-17 18:34:56 -07:00
|
|
|
remove_proc_entry("ipv6_route", net->proc_net);
|
|
|
|
remove_proc_entry("rt6_stats", net->proc_net);
|
2012-06-18 06:08:33 -06:00
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2008-03-04 14:45:33 -07:00
|
|
|
static struct pernet_operations ip6_route_net_ops = {
|
|
|
|
.init = ip6_route_net_init,
|
|
|
|
.exit = ip6_route_net_exit,
|
|
|
|
};
|
|
|
|
|
2012-06-09 17:27:05 -06:00
|
|
|
static int __net_init ipv6_inetpeer_init(struct net *net)
|
|
|
|
{
|
|
|
|
struct inet_peer_base *bp = kmalloc(sizeof(*bp), GFP_KERNEL);
|
|
|
|
|
|
|
|
if (!bp)
|
|
|
|
return -ENOMEM;
|
|
|
|
inet_peer_base_init(bp);
|
|
|
|
net->ipv6.peers = bp;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __net_exit ipv6_inetpeer_exit(struct net *net)
|
|
|
|
{
|
|
|
|
struct inet_peer_base *bp = net->ipv6.peers;
|
|
|
|
|
|
|
|
net->ipv6.peers = NULL;
|
2012-06-09 17:32:41 -06:00
|
|
|
inetpeer_invalidate_tree(bp);
|
2012-06-09 17:27:05 -06:00
|
|
|
kfree(bp);
|
|
|
|
}
|
|
|
|
|
2012-06-09 20:00:16 -06:00
|
|
|
static struct pernet_operations ipv6_inetpeer_ops = {
|
2012-06-09 17:27:05 -06:00
|
|
|
.init = ipv6_inetpeer_init,
|
|
|
|
.exit = ipv6_inetpeer_exit,
|
|
|
|
};
|
|
|
|
|
2012-06-18 06:08:33 -06:00
|
|
|
static struct pernet_operations ip6_route_net_late_ops = {
|
|
|
|
.init = ip6_route_net_init_late,
|
|
|
|
.exit = ip6_route_net_exit_late,
|
|
|
|
};
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
static struct notifier_block ip6_route_dev_notifier = {
|
|
|
|
.notifier_call = ip6_route_dev_notify,
|
|
|
|
.priority = 0,
|
|
|
|
};
|
|
|
|
|
2007-12-07 01:43:48 -07:00
|
|
|
int __init ip6_route_init(void)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2007-12-07 01:43:48 -07:00
|
|
|
int ret;
|
|
|
|
|
2008-03-04 14:48:53 -07:00
|
|
|
ret = -ENOMEM;
|
|
|
|
ip6_dst_ops_template.kmem_cachep =
|
2006-08-26 20:25:52 -06:00
|
|
|
kmem_cache_create("ip6_dst_cache", sizeof(struct rt6_info), 0,
|
2007-12-07 01:45:16 -07:00
|
|
|
SLAB_HWCACHE_ALIGN, NULL);
|
2008-03-04 14:48:53 -07:00
|
|
|
if (!ip6_dst_ops_template.kmem_cachep)
|
2009-01-07 19:09:08 -07:00
|
|
|
goto out;
|
2007-05-24 19:17:54 -06:00
|
|
|
|
2010-10-08 00:37:34 -06:00
|
|
|
ret = dst_entries_init(&ip6_dst_blackhole_ops);
|
2008-03-04 14:48:30 -07:00
|
|
|
if (ret)
|
2008-03-04 14:48:10 -07:00
|
|
|
goto out_kmem_cache;
|
|
|
|
|
2012-06-09 17:27:05 -06:00
|
|
|
ret = register_pernet_subsys(&ipv6_inetpeer_ops);
|
|
|
|
if (ret)
|
2012-06-16 02:12:19 -06:00
|
|
|
goto out_dst_entries;
|
2012-06-14 17:00:17 -06:00
|
|
|
|
2012-06-15 16:51:55 -06:00
|
|
|
ret = register_pernet_subsys(&ip6_route_net_ops);
|
|
|
|
if (ret)
|
|
|
|
goto out_register_inetpeer;
|
2012-06-09 17:27:05 -06:00
|
|
|
|
2008-10-01 03:37:56 -06:00
|
|
|
ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops_template.kmem_cachep;
|
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
/* Registering of the loopback is done before this portion of code,
|
|
|
|
* the loopback reference in rt6_info will not be taken, do it
|
|
|
|
* manually for init_net */
|
2010-06-11 00:31:35 -06:00
|
|
|
init_net.ipv6.ip6_null_entry->dst.dev = init_net.loopback_dev;
|
2008-03-04 14:48:30 -07:00
|
|
|
init_net.ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
|
|
|
|
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
|
2010-06-11 00:31:35 -06:00
|
|
|
init_net.ipv6.ip6_prohibit_entry->dst.dev = init_net.loopback_dev;
|
2008-03-04 14:48:30 -07:00
|
|
|
init_net.ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
|
2010-06-11 00:31:35 -06:00
|
|
|
init_net.ipv6.ip6_blk_hole_entry->dst.dev = init_net.loopback_dev;
|
2008-03-04 14:48:30 -07:00
|
|
|
init_net.ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
|
|
|
|
#endif
|
2012-06-16 02:12:19 -06:00
|
|
|
ret = fib6_init();
|
2007-12-07 01:43:48 -07:00
|
|
|
if (ret)
|
2008-03-04 14:48:30 -07:00
|
|
|
goto out_register_subsys;
|
2007-12-07 01:43:48 -07:00
|
|
|
|
|
|
|
ret = xfrm6_init();
|
|
|
|
if (ret)
|
2012-06-16 02:12:19 -06:00
|
|
|
goto out_fib6_init;
|
2007-12-08 01:14:11 -07:00
|
|
|
|
2007-12-07 01:43:48 -07:00
|
|
|
ret = fib6_rules_init();
|
|
|
|
if (ret)
|
|
|
|
goto xfrm6_init;
|
2007-12-08 01:14:54 -07:00
|
|
|
|
2012-06-18 06:08:33 -06:00
|
|
|
ret = register_pernet_subsys(&ip6_route_net_late_ops);
|
|
|
|
if (ret)
|
|
|
|
goto fib6_rules_init;
|
|
|
|
|
2007-12-07 01:43:48 -07:00
|
|
|
ret = -ENOBUFS;
|
2011-06-09 19:27:09 -06:00
|
|
|
if (__rtnl_register(PF_INET6, RTM_NEWROUTE, inet6_rtm_newroute, NULL, NULL) ||
|
|
|
|
__rtnl_register(PF_INET6, RTM_DELROUTE, inet6_rtm_delroute, NULL, NULL) ||
|
|
|
|
__rtnl_register(PF_INET6, RTM_GETROUTE, inet6_rtm_getroute, NULL, NULL))
|
2012-06-18 06:08:33 -06:00
|
|
|
goto out_register_late_subsys;
|
2007-03-22 12:58:32 -06:00
|
|
|
|
2008-03-04 14:48:30 -07:00
|
|
|
ret = register_netdevice_notifier(&ip6_route_dev_notifier);
|
2008-03-04 14:45:33 -07:00
|
|
|
if (ret)
|
2012-06-18 06:08:33 -06:00
|
|
|
goto out_register_late_subsys;
|
2008-03-04 14:48:30 -07:00
|
|
|
|
2007-12-07 01:43:48 -07:00
|
|
|
out:
|
|
|
|
return ret;
|
|
|
|
|
2012-06-18 06:08:33 -06:00
|
|
|
out_register_late_subsys:
|
|
|
|
unregister_pernet_subsys(&ip6_route_net_late_ops);
|
2007-12-07 01:43:48 -07:00
|
|
|
fib6_rules_init:
|
|
|
|
fib6_rules_cleanup();
|
|
|
|
xfrm6_init:
|
|
|
|
xfrm6_fini();
|
2012-06-14 17:00:17 -06:00
|
|
|
out_fib6_init:
|
|
|
|
fib6_gc_cleanup();
|
2008-03-04 14:48:30 -07:00
|
|
|
out_register_subsys:
|
|
|
|
unregister_pernet_subsys(&ip6_route_net_ops);
|
2012-06-15 16:51:55 -06:00
|
|
|
out_register_inetpeer:
|
|
|
|
unregister_pernet_subsys(&ipv6_inetpeer_ops);
|
2010-10-08 00:37:34 -06:00
|
|
|
out_dst_entries:
|
|
|
|
dst_entries_destroy(&ip6_dst_blackhole_ops);
|
2007-12-07 01:43:48 -07:00
|
|
|
out_kmem_cache:
|
2008-03-04 14:49:23 -07:00
|
|
|
kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep);
|
2007-12-07 01:43:48 -07:00
|
|
|
goto out;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
void ip6_route_cleanup(void)
|
|
|
|
{
|
2008-03-04 14:48:30 -07:00
|
|
|
unregister_netdevice_notifier(&ip6_route_dev_notifier);
|
2012-06-18 06:08:33 -06:00
|
|
|
unregister_pernet_subsys(&ip6_route_net_late_ops);
|
2006-08-04 04:39:02 -06:00
|
|
|
fib6_rules_cleanup();
|
2005-04-16 16:20:36 -06:00
|
|
|
xfrm6_fini();
|
|
|
|
fib6_gc_cleanup();
|
2012-06-09 17:27:05 -06:00
|
|
|
unregister_pernet_subsys(&ipv6_inetpeer_ops);
|
2008-03-04 14:48:30 -07:00
|
|
|
unregister_pernet_subsys(&ip6_route_net_ops);
|
net dst: fix percpu_counter list corruption and poison overwritten
There're some percpu_counter list corruption and poison overwritten warnings
in recent kernel, which is resulted by fc66f95c.
commit fc66f95c switches to use percpu_counter, in ip6_route_net_init, kernel
init the percpu_counter for dst entries, but, the percpu_counter is never destroyed
in ip6_route_net_exit. So if the related data is freed by kernel, the freed percpu_counter
is still on the list, then if we insert/remove other percpu_counter, list corruption
resulted. Also, if the insert/remove option modifies the ->prev,->next pointer of
the freed value, the poison overwritten is resulted then.
With the following patch, the percpu_counter list corruption and poison overwritten
warnings disappeared.
Signed-off-by: Xiaotian Feng <dfeng@redhat.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: "Pekka Savola (ipv6)" <pekkas@netcore.fi>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-11-02 10:11:05 -06:00
|
|
|
dst_entries_destroy(&ip6_dst_blackhole_ops);
|
2008-03-04 14:49:23 -07:00
|
|
|
kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|