linux/net/ipv6/route.c

3769 lines
90 KiB
C
Raw Normal View History

/*
* Linux INET6 implementation
* FIB front-end.
*
* Authors:
* Pedro Roque <roque@di.fc.ul.pt>
*
* 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.
* Ville Nuorvala
* Fixed routing subtrees.
*/
#define pr_fmt(fmt) "IPv6: " fmt
#include <linux/capability.h>
#include <linux/errno.h>
#include <linux/export.h>
#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>
#include <linux/mroute6.h>
#include <linux/init.h>
#include <linux/if_arp.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#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 08:04:11 +00:00
#include <linux/slab.h>
#include <net/net_namespace.h>
#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/dst_metadata.h>
#include <net/xfrm.h>
#include <net/netevent.h>
#include <net/netlink.h>
#include <net/nexthop.h>
#include <net/lwtunnel.h>
#include <net/ip_tunnels.h>
#include <asm/uaccess.h>
#ifdef CONFIG_SYSCTL
#include <linux/sysctl.h>
#endif
enum rt6_nud_state {
RT6_NUD_FAIL_HARD = -3,
RT6_NUD_FAIL_PROBE = -2,
RT6_NUD_FAIL_DO_RR = -1,
RT6_NUD_SUCCEED = 1
};
static void ip6_rt_copy_init(struct rt6_info *rt, struct rt6_info *ort);
static struct dst_entry *ip6_dst_check(struct dst_entry *dst, u32 cookie);
static unsigned int ip6_default_advmss(const struct dst_entry *dst);
static unsigned int ip6_mtu(const struct dst_entry *dst);
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);
static int ip6_dst_gc(struct dst_ops *ops);
static int ip6_pkt_discard(struct sk_buff *skb);
static int ip6_pkt_discard_out(struct sock *sk, struct sk_buff *skb);
static int ip6_pkt_prohibit(struct sk_buff *skb);
static int ip6_pkt_prohibit_out(struct sock *sk, struct sk_buff *skb);
static void ip6_link_failure(struct sk_buff *skb);
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);
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
static void rt6_dst_from_metrics_check(struct rt6_info *rt);
static int rt6_score_route(struct rt6_info *rt, int oif, int strict);
#ifdef CONFIG_IPV6_ROUTE_INFO
static struct rt6_info *rt6_add_route_info(struct net *net,
const struct in6_addr *prefix, int prefixlen,
const struct in6_addr *gwaddr, int ifindex,
unsigned int pref);
static struct rt6_info *rt6_get_route_info(struct net *net,
const struct in6_addr *prefix, int prefixlen,
const struct in6_addr *gwaddr, int ifindex);
#endif
struct uncached_list {
spinlock_t lock;
struct list_head head;
};
static DEFINE_PER_CPU_ALIGNED(struct uncached_list, rt6_uncached_list);
static void rt6_uncached_list_add(struct rt6_info *rt)
{
struct uncached_list *ul = raw_cpu_ptr(&rt6_uncached_list);
rt->dst.flags |= DST_NOCACHE;
rt->rt6i_uncached_list = ul;
spin_lock_bh(&ul->lock);
list_add_tail(&rt->rt6i_uncached, &ul->head);
spin_unlock_bh(&ul->lock);
}
static void rt6_uncached_list_del(struct rt6_info *rt)
{
if (!list_empty(&rt->rt6i_uncached)) {
struct uncached_list *ul = rt->rt6i_uncached_list;
spin_lock_bh(&ul->lock);
list_del(&rt->rt6i_uncached);
spin_unlock_bh(&ul->lock);
}
}
static void rt6_uncached_list_flush_dev(struct net *net, struct net_device *dev)
{
struct net_device *loopback_dev = net->loopback_dev;
int cpu;
if (dev == loopback_dev)
return;
for_each_possible_cpu(cpu) {
struct uncached_list *ul = per_cpu_ptr(&rt6_uncached_list, cpu);
struct rt6_info *rt;
spin_lock_bh(&ul->lock);
list_for_each_entry(rt, &ul->head, rt6i_uncached) {
struct inet6_dev *rt_idev = rt->rt6i_idev;
struct net_device *rt_dev = rt->dst.dev;
if (rt_idev->dev == dev) {
rt->rt6i_idev = in6_dev_get(loopback_dev);
in6_dev_put(rt_idev);
}
if (rt_dev == dev) {
rt->dst.dev = loopback_dev;
dev_hold(rt->dst.dev);
dev_put(rt_dev);
}
}
spin_unlock_bh(&ul->lock);
}
}
static u32 *rt6_pcpu_cow_metrics(struct rt6_info *rt)
{
return dst_metrics_write_ptr(rt->dst.from);
}
static u32 *ipv6_cow_metrics(struct dst_entry *dst, unsigned long old)
{
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
struct rt6_info *rt = (struct rt6_info *)dst;
if (rt->rt6i_flags & RTF_PCPU)
return rt6_pcpu_cow_metrics(rt);
else if (rt->rt6i_flags & RTF_CACHE)
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
return NULL;
else
return dst_cow_metrics_generic(dst, old);
}
static inline const void *choose_neigh_daddr(struct rt6_info *rt,
struct sk_buff *skb,
const void *daddr)
{
struct in6_addr *p = &rt->rt6i_gateway;
if (!ipv6_addr_any(p))
return (const void *) p;
else if (skb)
return &ipv6_hdr(skb)->daddr;
return daddr;
}
static struct neighbour *ip6_neigh_lookup(const struct dst_entry *dst,
struct sk_buff *skb,
const void *daddr)
{
struct rt6_info *rt = (struct rt6_info *) dst;
struct neighbour *n;
daddr = choose_neigh_daddr(rt, skb, daddr);
n = __ipv6_neigh_lookup(dst->dev, daddr);
if (n)
return n;
return neigh_create(&nd_tbl, daddr, dst->dev);
}
static struct dst_ops ip6_dst_ops_template = {
.family = AF_INET6,
.gc = ip6_dst_gc,
.gc_thresh = 1024,
.check = ip6_dst_check,
.default_advmss = ip6_default_advmss,
.mtu = ip6_mtu,
.cow_metrics = ipv6_cow_metrics,
.destroy = ip6_dst_destroy,
.ifdown = ip6_dst_ifdown,
.negative_advice = ip6_negative_advice,
.link_failure = ip6_link_failure,
.update_pmtu = ip6_rt_update_pmtu,
.redirect = rt6_do_redirect,
.local_out = __ip6_local_out,
.neigh_lookup = ip6_neigh_lookup,
};
static unsigned int ip6_blackhole_mtu(const struct dst_entry *dst)
{
unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
return mtu ? : dst->dev->mtu;
}
static void ip6_rt_blackhole_update_pmtu(struct dst_entry *dst, struct sock *sk,
struct sk_buff *skb, u32 mtu)
{
}
static void ip6_rt_blackhole_redirect(struct dst_entry *dst, struct sock *sk,
struct sk_buff *skb)
{
}
static struct dst_ops ip6_dst_blackhole_ops = {
.family = AF_INET6,
.destroy = ip6_dst_destroy,
.check = ip6_dst_check,
.mtu = ip6_blackhole_mtu,
.default_advmss = ip6_default_advmss,
.update_pmtu = ip6_rt_blackhole_update_pmtu,
.redirect = ip6_rt_blackhole_redirect,
.cow_metrics = dst_cow_metrics_generic,
.neigh_lookup = ip6_neigh_lookup,
};
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>
2011-01-27 04:51:05 +00:00
static const u32 ip6_template_metrics[RTAX_MAX] = {
[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>
2011-01-27 04:51:05 +00:00
};
static const struct rt6_info ip6_null_entry_template = {
.dst = {
.__refcnt = ATOMIC_INIT(1),
.__use = 1,
.obsolete = DST_OBSOLETE_FORCE_CHK,
.error = -ENETUNREACH,
.input = ip6_pkt_discard,
.output = ip6_pkt_discard_out,
},
.rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP),
.rt6i_protocol = RTPROT_KERNEL,
.rt6i_metric = ~(u32) 0,
.rt6i_ref = ATOMIC_INIT(1),
};
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
static const struct rt6_info ip6_prohibit_entry_template = {
.dst = {
.__refcnt = ATOMIC_INIT(1),
.__use = 1,
.obsolete = DST_OBSOLETE_FORCE_CHK,
.error = -EACCES,
.input = ip6_pkt_prohibit,
.output = ip6_pkt_prohibit_out,
},
.rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP),
.rt6i_protocol = RTPROT_KERNEL,
.rt6i_metric = ~(u32) 0,
.rt6i_ref = ATOMIC_INIT(1),
};
static const struct rt6_info ip6_blk_hole_entry_template = {
.dst = {
.__refcnt = ATOMIC_INIT(1),
.__use = 1,
.obsolete = DST_OBSOLETE_FORCE_CHK,
.error = -EINVAL,
.input = dst_discard,
.output = dst_discard_sk,
},
.rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP),
.rt6i_protocol = RTPROT_KERNEL,
.rt6i_metric = ~(u32) 0,
.rt6i_ref = ATOMIC_INIT(1),
};
#endif
static void rt6_info_init(struct rt6_info *rt)
{
struct dst_entry *dst = &rt->dst;
memset(dst + 1, 0, sizeof(*rt) - sizeof(*dst));
INIT_LIST_HEAD(&rt->rt6i_siblings);
INIT_LIST_HEAD(&rt->rt6i_uncached);
}
/* allocate dst with ip6_dst_ops */
static struct rt6_info *__ip6_dst_alloc(struct net *net,
struct net_device *dev,
int flags)
{
struct rt6_info *rt = dst_alloc(&net->ipv6.ip6_dst_ops, dev,
0, DST_OBSOLETE_FORCE_CHK, flags);
if (rt)
rt6_info_init(rt);
return rt;
}
static struct rt6_info *ip6_dst_alloc(struct net *net,
struct net_device *dev,
int flags)
{
struct rt6_info *rt = __ip6_dst_alloc(net, dev, flags);
if (rt) {
rt->rt6i_pcpu = alloc_percpu_gfp(struct rt6_info *, GFP_ATOMIC);
if (rt->rt6i_pcpu) {
int cpu;
for_each_possible_cpu(cpu) {
struct rt6_info **p;
p = per_cpu_ptr(rt->rt6i_pcpu, cpu);
/* no one shares rt */
*p = NULL;
}
} else {
dst_destroy((struct dst_entry *)rt);
return NULL;
}
}
return rt;
}
static void ip6_dst_destroy(struct dst_entry *dst)
{
struct rt6_info *rt = (struct rt6_info *)dst;
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>
2013-02-20 00:29:08 +00:00
struct dst_entry *from = dst->from;
struct inet6_dev *idev;
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
dst_destroy_metrics_generic(dst);
free_percpu(rt->rt6i_pcpu);
rt6_uncached_list_del(rt);
idev = rt->rt6i_idev;
if (idev) {
rt->rt6i_idev = NULL;
in6_dev_put(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>
2013-02-20 00:29:08 +00:00
dst->from = NULL;
dst_release(from);
}
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;
struct net_device *loopback_dev =
dev_net(dev)->loopback_dev;
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);
}
}
}
}
static bool rt6_check_expired(const struct rt6_info *rt)
{
if (rt->rt6i_flags & RTF_EXPIRES) {
if (time_after(jiffies, rt->dst.expires))
return true;
} else if (rt->dst.from) {
return rt6_check_expired((struct rt6_info *) rt->dst.from);
}
return false;
}
/* 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;
val ^= ipv6_addr_hash(&fl6->daddr);
val ^= ipv6_addr_hash(&fl6->saddr);
/* Work only if this not encapsulated */
switch (fl6->flowi6_proto) {
case IPPROTO_UDP:
case IPPROTO_TCP:
case IPPROTO_SCTP:
val ^= (__force u16)fl6->fl6_sport;
val ^= (__force u16)fl6->fl6_dport;
break;
case IPPROTO_ICMPV6:
val ^= (__force u16)fl6->fl6_icmp_type;
val ^= (__force u16)fl6->fl6_icmp_code;
break;
}
/* RFC6438 recommands to use flowlabel */
val ^= (__force u32)fl6->flowlabel;
/* 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,
struct flowi6 *fl6, int oif,
int strict)
{
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) {
if (rt6_score_route(sibling, oif, strict) < 0)
break;
match = sibling;
break;
}
}
return match;
}
/*
* Route lookup. Any table->tb6_lock is implied.
*/
static inline struct rt6_info *rt6_device_match(struct net *net,
struct rt6_info *rt,
const struct in6_addr *saddr,
int oif,
int flags)
{
struct rt6_info *local = NULL;
struct rt6_info *sprt;
if (!oif && ipv6_addr_any(saddr))
goto out;
for (sprt = rt; sprt; sprt = sprt->dst.rt6_next) {
struct net_device *dev = sprt->dst.dev;
if (oif) {
if (dev->ifindex == oif)
return sprt;
if (dev->flags & IFF_LOOPBACK) {
if (!sprt->rt6i_idev ||
sprt->rt6i_idev->dev->ifindex != oif) {
if (flags & RT6_LOOKUP_F_IFACE && oif)
continue;
if (local && (!oif ||
local->rt6i_idev->dev->ifindex == oif))
continue;
}
local = sprt;
}
} else {
if (ipv6_chk_addr(net, saddr, dev,
flags & RT6_LOOKUP_F_IFACE))
return sprt;
}
}
if (oif) {
if (local)
return local;
if (flags & RT6_LOOKUP_F_IFACE)
return net->ipv6.ip6_null_entry;
}
out:
return rt;
}
#ifdef CONFIG_IPV6_ROUTER_PREF
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, NULL);
dev_put(work->dev);
kfree(work);
}
static void rt6_probe(struct rt6_info *rt)
{
struct __rt6_probe_work *work;
struct neighbour *neigh;
/*
* 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.
*/
if (!rt || !(rt->rt6i_flags & RTF_GATEWAY))
return;
rcu_read_lock_bh();
neigh = __ipv6_neigh_lookup_noref(rt->dst.dev, &rt->rt6i_gateway);
if (neigh) {
if (neigh->nud_state & NUD_VALID)
goto out;
work = NULL;
write_lock(&neigh->lock);
if (!(neigh->nud_state & NUD_VALID) &&
time_after(jiffies,
neigh->updated +
rt->rt6i_idev->cnf.rtr_probe_interval)) {
work = kmalloc(sizeof(*work), GFP_ATOMIC);
if (work)
__neigh_set_probe_once(neigh);
}
write_unlock(&neigh->lock);
} else {
work = kmalloc(sizeof(*work), GFP_ATOMIC);
}
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);
}
out:
rcu_read_unlock_bh();
}
#else
static inline void rt6_probe(struct rt6_info *rt)
{
}
#endif
/*
* Default Router Selection (RFC 2461 6.3.6)
*/
static inline int rt6_check_dev(struct rt6_info *rt, int oif)
{
struct net_device *dev = rt->dst.dev;
if (!oif || dev->ifindex == oif)
return 2;
if ((dev->flags & IFF_LOOPBACK) &&
rt->rt6i_idev && rt->rt6i_idev->dev->ifindex == oif)
return 1;
return 0;
}
static inline enum rt6_nud_state rt6_check_neigh(struct rt6_info *rt)
{
struct neighbour *neigh;
enum rt6_nud_state ret = RT6_NUD_FAIL_HARD;
if (rt->rt6i_flags & RTF_NONEXTHOP ||
!(rt->rt6i_flags & RTF_GATEWAY))
return RT6_NUD_SUCCEED;
rcu_read_lock_bh();
neigh = __ipv6_neigh_lookup_noref(rt->dst.dev, &rt->rt6i_gateway);
if (neigh) {
read_lock(&neigh->lock);
if (neigh->nud_state & NUD_VALID)
ret = RT6_NUD_SUCCEED;
#ifdef CONFIG_IPV6_ROUTER_PREF
else if (!(neigh->nud_state & NUD_FAILED))
ret = RT6_NUD_SUCCEED;
else
ret = RT6_NUD_FAIL_PROBE;
#endif
read_unlock(&neigh->lock);
} else {
ret = IS_ENABLED(CONFIG_IPV6_ROUTER_PREF) ?
RT6_NUD_SUCCEED : RT6_NUD_FAIL_DO_RR;
}
rcu_read_unlock_bh();
return ret;
}
static int rt6_score_route(struct rt6_info *rt, int oif,
int strict)
{
int m;
m = rt6_check_dev(rt, oif);
if (!m && (strict & RT6_LOOKUP_F_IFACE))
return RT6_NUD_FAIL_HARD;
#ifdef CONFIG_IPV6_ROUTER_PREF
m |= IPV6_DECODE_PREF(IPV6_EXTRACT_PREF(rt->rt6i_flags)) << 2;
#endif
if (strict & RT6_LOOKUP_F_REACHABLE) {
int n = rt6_check_neigh(rt);
if (n < 0)
return n;
}
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>
2007-03-25 03:36:25 +00:00
static struct rt6_info *find_match(struct rt6_info *rt, int oif, int strict,
int *mpri, struct rt6_info *match,
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>
2007-03-25 03:36:25 +00:00
int m;
bool match_do_rr = false;
net: ipv6 sysctl option to ignore routes when nexthop link is down Like the ipv4 patch with a similar title, this adds a sysctl to allow the user to change routing behavior based on whether or not the interface associated with the nexthop was an up or down link. The default setting preserves the current behavior, but anyone that enables it will notice that nexthops on down interfaces will no longer be selected: net.ipv6.conf.all.ignore_routes_with_linkdown = 0 net.ipv6.conf.default.ignore_routes_with_linkdown = 0 net.ipv6.conf.lo.ignore_routes_with_linkdown = 0 ... When the above sysctls are set, not only will link status be reported to userspace, but an indication that a nexthop is dead and will not be used is also reported. 1000::/8 via 7000::2 dev p7p1 metric 1024 dead linkdown pref medium 1000::/8 via 8000::2 dev p8p1 metric 1024 pref medium 7000::/8 dev p7p1 proto kernel metric 256 dead linkdown pref medium 8000::/8 dev p8p1 proto kernel metric 256 pref medium 9000::/8 via 8000::2 dev p8p1 metric 2048 pref medium 9000::/8 via 7000::2 dev p7p1 metric 1024 dead linkdown pref medium fe80::/64 dev p7p1 proto kernel metric 256 dead linkdown pref medium fe80::/64 dev p8p1 proto kernel metric 256 pref medium This also adds devconf support and notification when sysctl values change. v2: drop use of rt6i_nhflags since it is not needed right now Signed-off-by: Andy Gospodarek <gospo@cumulusnetworks.com> Signed-off-by: Dinesh Dutt <ddutt@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-13 14:39:01 +00:00
struct inet6_dev *idev = rt->rt6i_idev;
struct net_device *dev = rt->dst.dev;
if (dev && !netif_carrier_ok(dev) &&
idev->cnf.ignore_routes_with_linkdown)
goto out;
[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>
2007-03-25 03:36:25 +00:00
if (rt6_check_expired(rt))
goto out;
m = rt6_score_route(rt, oif, strict);
if (m == RT6_NUD_FAIL_DO_RR) {
match_do_rr = true;
m = 0; /* lowest valid score */
} else if (m == RT6_NUD_FAIL_HARD) {
[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>
2007-03-25 03:36:25 +00:00
goto out;
}
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>
2007-03-25 03:36:25 +00:00
/* note that m can be RT6_NUD_FAIL_PROBE at this point */
[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>
2007-03-25 03:36:25 +00:00
if (m > *mpri) {
*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>
2007-03-25 03:36:25 +00:00
*mpri = m;
match = rt;
}
out:
return match;
}
static struct rt6_info *find_rr_leaf(struct fib6_node *fn,
struct rt6_info *rr_head,
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>
2007-03-25 03:36:25 +00:00
{
struct rt6_info *rt, *match, *cont;
int mpri = -1;
[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>
2007-03-25 03:36:25 +00:00
match = NULL;
cont = NULL;
for (rt = rr_head; rt; rt = rt->dst.rt6_next) {
if (rt->rt6i_metric != metric) {
cont = rt;
break;
}
match = find_match(rt, oif, strict, &mpri, match, do_rr);
}
for (rt = fn->leaf; rt && rt != rr_head; rt = rt->dst.rt6_next) {
if (rt->rt6i_metric != metric) {
cont = rt;
break;
}
match = find_match(rt, oif, strict, &mpri, match, do_rr);
}
if (match || !cont)
return match;
for (rt = cont; rt; rt = rt->dst.rt6_next)
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>
2007-03-25 03:36:25 +00:00
return match;
}
[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>
2007-03-25 03:36:25 +00:00
static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
{
struct rt6_info *match, *rt0;
struct net *net;
bool 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>
2007-03-25 03:36:25 +00:00
rt0 = fn->rr_ptr;
if (!rt0)
fn->rr_ptr = rt0 = fn->leaf;
match = find_rr_leaf(fn, rt0, rt0->rt6i_metric, oif, strict,
&do_rr);
if (do_rr) {
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>
2007-03-25 03:36:25 +00:00
/* 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>
2007-03-25 03:36:25 +00:00
if (!next || next->rt6i_metric != rt0->rt6i_metric)
next = fn->leaf;
if (next != rt0)
fn->rr_ptr = next;
}
net = dev_net(rt0->dst.dev);
return match ? match : net->ipv6.ip6_null_entry;
}
static bool rt6_is_gw_or_nonexthop(const struct rt6_info *rt)
{
return (rt->rt6i_flags & (RTF_NONEXTHOP | RTF_GATEWAY));
}
#ifdef CONFIG_IPV6_ROUTE_INFO
int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
const struct in6_addr *gwaddr)
{
struct net *net = dev_net(dev);
struct route_info *rinfo = (struct route_info *) opt;
struct in6_addr prefix_buf, *prefix;
unsigned int pref;
unsigned long lifetime;
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)
return -EINVAL;
lifetime = addrconf_timeout_fixup(ntohl(rinfo->lifetime), HZ);
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;
}
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);
if (rt && !lifetime) {
ip6_del_rt(rt);
rt = NULL;
}
if (!rt && lifetime)
rt = rt6_add_route_info(net, prefix, rinfo->prefix_len, gwaddr, dev->ifindex,
pref);
else if (rt)
rt->rt6i_flags = RTF_ROUTEINFO |
(rt->rt6i_flags & ~RTF_PREF_MASK) | RTF_PREF(pref);
if (rt) {
if (!addrconf_finite_timeout(lifetime))
rt6_clean_expires(rt);
else
rt6_set_expires(rt, jiffies + HZ * lifetime);
ip6_rt_put(rt);
}
return 0;
}
#endif
static struct fib6_node* fib6_backtrack(struct fib6_node *fn,
struct in6_addr *saddr)
{
struct fib6_node *pn;
while (1) {
if (fn->fn_flags & RTN_TL_ROOT)
return NULL;
pn = fn->parent;
if (FIB6_SUBTREE(pn) && FIB6_SUBTREE(pn) != fn)
fn = fib6_lookup(FIB6_SUBTREE(pn), NULL, saddr);
else
fn = pn;
if (fn->fn_flags & RTN_RTINFO)
return fn;
}
}
static struct rt6_info *ip6_pol_route_lookup(struct net *net,
struct fib6_table *table,
struct flowi6 *fl6, int flags)
{
struct fib6_node *fn;
struct rt6_info *rt;
read_lock_bh(&table->tb6_lock);
fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr);
restart:
rt = fn->leaf;
rt = rt6_device_match(net, rt, &fl6->saddr, fl6->flowi6_oif, flags);
if (rt->rt6i_nsiblings && fl6->flowi6_oif == 0)
rt = rt6_multipath_select(rt, fl6, fl6->flowi6_oif, flags);
if (rt == net->ipv6.ip6_null_entry) {
fn = fib6_backtrack(fn, &fl6->saddr);
if (fn)
goto restart;
}
dst_use(&rt->dst, jiffies);
read_unlock_bh(&table->tb6_lock);
return rt;
}
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);
struct rt6_info *rt6_lookup(struct net *net, const struct in6_addr *daddr,
const struct in6_addr *saddr, int oif, int strict)
{
struct flowi6 fl6 = {
.flowi6_oif = oif,
.daddr = *daddr,
};
struct dst_entry *dst;
int flags = strict ? RT6_LOOKUP_F_IFACE : 0;
if (saddr) {
memcpy(&fl6.saddr, saddr, sizeof(*saddr));
flags |= RT6_LOOKUP_F_HAS_SADDR;
}
dst = fib6_rule_lookup(net, &fl6, flags, ip6_pol_route_lookup);
if (dst->error == 0)
return (struct rt6_info *) dst;
dst_release(dst);
return NULL;
}
EXPORT_SYMBOL(rt6_lookup);
/* ip6_ins_rt is called with FREE table->tb6_lock.
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.
*/
ipv6: do not overwrite inetpeer metrics prematurely If an IPv6 host route with metrics exists, an attempt to add a new route for the same target with different metrics fails but rewrites the metrics anyway: 12sp0:~ # ip route add fec0::1 dev eth0 rto_min 1000 12sp0:~ # ip -6 route show fe80::/64 dev eth0 proto kernel metric 256 fec0::1 dev eth0 metric 1024 rto_min lock 1s 12sp0:~ # ip route add fec0::1 dev eth0 rto_min 1500 RTNETLINK answers: File exists 12sp0:~ # ip -6 route show fe80::/64 dev eth0 proto kernel metric 256 fec0::1 dev eth0 metric 1024 rto_min lock 1.5s This is caused by all IPv6 host routes using the metrics in their inetpeer (or the shared default). This also holds for the new route created in ip6_route_add() which shares the metrics with the already existing route and thus ip6_route_add() rewrites the metrics even if the new route ends up not being used at all. Another problem is that old metrics in inetpeer can reappear unexpectedly for a new route, e.g. 12sp0:~ # ip route add fec0::1 dev eth0 rto_min 1000 12sp0:~ # ip route del fec0::1 12sp0:~ # ip route add fec0::1 dev eth0 12sp0:~ # ip route change fec0::1 dev eth0 hoplimit 10 12sp0:~ # ip -6 route show fe80::/64 dev eth0 proto kernel metric 256 fec0::1 dev eth0 metric 1024 hoplimit 10 rto_min lock 1s Resolve the first problem by moving the setting of metrics down into fib6_add_rt2node() to the point we are sure we are inserting the new route into the tree. Second problem is addressed by introducing new flag DST_METRICS_FORCE_OVERWRITE which is set for a new host route in ip6_route_add() and makes ipv6_cow_metrics() always overwrite the metrics in inetpeer (even if they are not "new"); it is reset after that. v5: use a flag in _metrics member rather than one in flags v4: fix a typo making a condition always true (thanks to Hannes Frederic Sowa) v3: rewritten based on David Miller's idea to move setting the metrics (and allocation in non-host case) down to the point we already know the route is to be inserted. Also rebased to net-next as it is quite late in the cycle. Signed-off-by: Michal Kubecek <mkubecek@suse.cz> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-27 12:04:08 +00:00
static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info,
struct mx6_config *mxc)
{
int err;
struct fib6_table *table;
table = rt->rt6i_table;
write_lock_bh(&table->tb6_lock);
err = fib6_add(&table->tb6_root, rt, info, mxc);
write_unlock_bh(&table->tb6_lock);
return err;
}
int ip6_ins_rt(struct rt6_info *rt)
{
struct nl_info info = { .nl_net = dev_net(rt->dst.dev), };
struct mx6_config mxc = { .mx = NULL, };
return __ip6_ins_rt(rt, &info, &mxc);
}
static struct rt6_info *ip6_rt_cache_alloc(struct rt6_info *ort,
const struct in6_addr *daddr,
const struct in6_addr *saddr)
{
struct rt6_info *rt;
/*
* Clone the route.
*/
if (ort->rt6i_flags & (RTF_CACHE | RTF_PCPU))
ort = (struct rt6_info *)ort->dst.from;
rt = __ip6_dst_alloc(dev_net(ort->dst.dev), ort->dst.dev, 0);
if (!rt)
return NULL;
ip6_rt_copy_init(rt, ort);
rt->rt6i_flags |= RTF_CACHE;
rt->rt6i_metric = 0;
rt->dst.flags |= DST_HOST;
rt->rt6i_dst.addr = *daddr;
rt->rt6i_dst.plen = 128;
if (!rt6_is_gw_or_nonexthop(ort)) {
if (ort->rt6i_dst.plen != 128 &&
ipv6_addr_equal(&ort->rt6i_dst.addr, daddr))
rt->rt6i_flags |= RTF_ANYCAST;
#ifdef CONFIG_IPV6_SUBTREES
if (rt->rt6i_src.plen && saddr) {
rt->rt6i_src.addr = *saddr;
rt->rt6i_src.plen = 128;
}
#endif
}
return rt;
}
static struct rt6_info *ip6_rt_pcpu_alloc(struct rt6_info *rt)
{
struct rt6_info *pcpu_rt;
pcpu_rt = __ip6_dst_alloc(dev_net(rt->dst.dev),
rt->dst.dev, rt->dst.flags);
if (!pcpu_rt)
return NULL;
ip6_rt_copy_init(pcpu_rt, rt);
pcpu_rt->rt6i_protocol = rt->rt6i_protocol;
pcpu_rt->rt6i_flags |= RTF_PCPU;
return pcpu_rt;
}
/* It should be called with read_lock_bh(&tb6_lock) acquired */
static struct rt6_info *rt6_get_pcpu_route(struct rt6_info *rt)
{
struct rt6_info *pcpu_rt, **p;
p = this_cpu_ptr(rt->rt6i_pcpu);
pcpu_rt = *p;
if (pcpu_rt) {
dst_hold(&pcpu_rt->dst);
rt6_dst_from_metrics_check(pcpu_rt);
}
return pcpu_rt;
}
static struct rt6_info *rt6_make_pcpu_route(struct rt6_info *rt)
{
ipv6: Fix a potential deadlock when creating pcpu rt rt6_make_pcpu_route() is called under read_lock(&table->tb6_lock). rt6_make_pcpu_route() calls ip6_rt_pcpu_alloc(rt) which then calls dst_alloc(). dst_alloc() _may_ call ip6_dst_gc() which takes the write_lock(&tabl->tb6_lock). A visualized version: read_lock(&table->tb6_lock); rt6_make_pcpu_route(); => ip6_rt_pcpu_alloc(); => dst_alloc(); => ip6_dst_gc(); => write_lock(&table->tb6_lock); /* oops */ The fix is to do a read_unlock first before calling ip6_rt_pcpu_alloc(). A reported stack: [141625.537638] INFO: rcu_sched self-detected stall on CPU { 27} (t=60000 jiffies g=4159086 c=4159085 q=2139) [141625.547469] Task dump for CPU 27: [141625.550881] mtr R running task 0 22121 22081 0x00000008 [141625.558069] 0000000000000000 ffff88103f363d98 ffffffff8106e488 000000000000001b [141625.565641] ffffffff81684900 ffff88103f363db8 ffffffff810702b0 0000000008000000 [141625.573220] ffffffff81684900 ffff88103f363de8 ffffffff8108df9f ffff88103f375a00 [141625.580803] Call Trace: [141625.583345] <IRQ> [<ffffffff8106e488>] sched_show_task+0xc1/0xc6 [141625.589650] [<ffffffff810702b0>] dump_cpu_task+0x35/0x39 [141625.595144] [<ffffffff8108df9f>] rcu_dump_cpu_stacks+0x6a/0x8c [141625.601320] [<ffffffff81090606>] rcu_check_callbacks+0x1f6/0x5d4 [141625.607669] [<ffffffff810940c8>] update_process_times+0x2a/0x4f [141625.613925] [<ffffffff8109fbee>] tick_sched_handle+0x32/0x3e [141625.619923] [<ffffffff8109fc2f>] tick_sched_timer+0x35/0x5c [141625.625830] [<ffffffff81094a1f>] __hrtimer_run_queues+0x8f/0x18d [141625.632171] [<ffffffff81094c9e>] hrtimer_interrupt+0xa0/0x166 [141625.638258] [<ffffffff8102bf2a>] local_apic_timer_interrupt+0x4e/0x52 [141625.645036] [<ffffffff8102c36f>] smp_apic_timer_interrupt+0x39/0x4a [141625.651643] [<ffffffff8140b9e8>] apic_timer_interrupt+0x68/0x70 [141625.657895] <EOI> [<ffffffff81346ee8>] ? dst_destroy+0x7c/0xb5 [141625.664188] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.670272] [<ffffffff81082b45>] ? queue_write_lock_slowpath+0x60/0x6f [141625.677140] [<ffffffff8140aa33>] _raw_write_lock_bh+0x23/0x25 [141625.683218] [<ffffffff813d4553>] __fib6_clean_all+0x40/0x82 [141625.689124] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.695207] [<ffffffff813d6058>] fib6_clean_all+0xe/0x10 [141625.700854] [<ffffffff813d60d3>] fib6_run_gc+0x79/0xc8 [141625.706329] [<ffffffff813d0510>] ip6_dst_gc+0x85/0xf9 [141625.711718] [<ffffffff81346d68>] dst_alloc+0x55/0x159 [141625.717105] [<ffffffff813d09b5>] __ip6_dst_alloc.isra.32+0x19/0x63 [141625.723620] [<ffffffff813d1830>] ip6_pol_route+0x36a/0x3e8 [141625.729441] [<ffffffff813d18d6>] ip6_pol_route_output+0x11/0x13 [141625.735700] [<ffffffff813f02c8>] fib6_rule_action+0xa7/0x1bf [141625.741698] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.748043] [<ffffffff81357c48>] fib_rules_lookup+0xb5/0x12a [141625.754050] [<ffffffff81141628>] ? poll_select_copy_remaining+0xf9/0xf9 [141625.761002] [<ffffffff813f0535>] fib6_rule_lookup+0x37/0x5c [141625.766914] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.773260] [<ffffffff813d008c>] ip6_route_output+0x7a/0x82 [141625.779177] [<ffffffff813c44c8>] ip6_dst_lookup_tail+0x53/0x112 [141625.785437] [<ffffffff813c45c3>] ip6_dst_lookup_flow+0x2a/0x6b [141625.791604] [<ffffffff813ddaab>] rawv6_sendmsg+0x407/0x9b6 [141625.797423] [<ffffffff813d7914>] ? do_ipv6_setsockopt.isra.8+0xd87/0xde2 [141625.804464] [<ffffffff8139d4b4>] inet_sendmsg+0x57/0x8e [141625.810028] [<ffffffff81329ba3>] sock_sendmsg+0x2e/0x3c [141625.815588] [<ffffffff8132be57>] SyS_sendto+0xfe/0x143 [141625.821063] [<ffffffff813dd551>] ? rawv6_setsockopt+0x5e/0x67 [141625.827146] [<ffffffff8132c9f8>] ? sock_common_setsockopt+0xf/0x11 [141625.833660] [<ffffffff8132c08c>] ? SyS_setsockopt+0x81/0xa2 [141625.839565] [<ffffffff8140ac17>] entry_SYSCALL_64_fastpath+0x12/0x6a Fixes: d52d3997f843 ("pv6: Create percpu rt6_info") Signed-off-by: Martin KaFai Lau <kafai@fb.com> CC: Hannes Frederic Sowa <hannes@stressinduktion.org> Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-14 18:05:54 +00:00
struct fib6_table *table = rt->rt6i_table;
struct rt6_info *pcpu_rt, *prev, **p;
pcpu_rt = ip6_rt_pcpu_alloc(rt);
if (!pcpu_rt) {
struct net *net = dev_net(rt->dst.dev);
ipv6: Fix a potential deadlock when creating pcpu rt rt6_make_pcpu_route() is called under read_lock(&table->tb6_lock). rt6_make_pcpu_route() calls ip6_rt_pcpu_alloc(rt) which then calls dst_alloc(). dst_alloc() _may_ call ip6_dst_gc() which takes the write_lock(&tabl->tb6_lock). A visualized version: read_lock(&table->tb6_lock); rt6_make_pcpu_route(); => ip6_rt_pcpu_alloc(); => dst_alloc(); => ip6_dst_gc(); => write_lock(&table->tb6_lock); /* oops */ The fix is to do a read_unlock first before calling ip6_rt_pcpu_alloc(). A reported stack: [141625.537638] INFO: rcu_sched self-detected stall on CPU { 27} (t=60000 jiffies g=4159086 c=4159085 q=2139) [141625.547469] Task dump for CPU 27: [141625.550881] mtr R running task 0 22121 22081 0x00000008 [141625.558069] 0000000000000000 ffff88103f363d98 ffffffff8106e488 000000000000001b [141625.565641] ffffffff81684900 ffff88103f363db8 ffffffff810702b0 0000000008000000 [141625.573220] ffffffff81684900 ffff88103f363de8 ffffffff8108df9f ffff88103f375a00 [141625.580803] Call Trace: [141625.583345] <IRQ> [<ffffffff8106e488>] sched_show_task+0xc1/0xc6 [141625.589650] [<ffffffff810702b0>] dump_cpu_task+0x35/0x39 [141625.595144] [<ffffffff8108df9f>] rcu_dump_cpu_stacks+0x6a/0x8c [141625.601320] [<ffffffff81090606>] rcu_check_callbacks+0x1f6/0x5d4 [141625.607669] [<ffffffff810940c8>] update_process_times+0x2a/0x4f [141625.613925] [<ffffffff8109fbee>] tick_sched_handle+0x32/0x3e [141625.619923] [<ffffffff8109fc2f>] tick_sched_timer+0x35/0x5c [141625.625830] [<ffffffff81094a1f>] __hrtimer_run_queues+0x8f/0x18d [141625.632171] [<ffffffff81094c9e>] hrtimer_interrupt+0xa0/0x166 [141625.638258] [<ffffffff8102bf2a>] local_apic_timer_interrupt+0x4e/0x52 [141625.645036] [<ffffffff8102c36f>] smp_apic_timer_interrupt+0x39/0x4a [141625.651643] [<ffffffff8140b9e8>] apic_timer_interrupt+0x68/0x70 [141625.657895] <EOI> [<ffffffff81346ee8>] ? dst_destroy+0x7c/0xb5 [141625.664188] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.670272] [<ffffffff81082b45>] ? queue_write_lock_slowpath+0x60/0x6f [141625.677140] [<ffffffff8140aa33>] _raw_write_lock_bh+0x23/0x25 [141625.683218] [<ffffffff813d4553>] __fib6_clean_all+0x40/0x82 [141625.689124] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.695207] [<ffffffff813d6058>] fib6_clean_all+0xe/0x10 [141625.700854] [<ffffffff813d60d3>] fib6_run_gc+0x79/0xc8 [141625.706329] [<ffffffff813d0510>] ip6_dst_gc+0x85/0xf9 [141625.711718] [<ffffffff81346d68>] dst_alloc+0x55/0x159 [141625.717105] [<ffffffff813d09b5>] __ip6_dst_alloc.isra.32+0x19/0x63 [141625.723620] [<ffffffff813d1830>] ip6_pol_route+0x36a/0x3e8 [141625.729441] [<ffffffff813d18d6>] ip6_pol_route_output+0x11/0x13 [141625.735700] [<ffffffff813f02c8>] fib6_rule_action+0xa7/0x1bf [141625.741698] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.748043] [<ffffffff81357c48>] fib_rules_lookup+0xb5/0x12a [141625.754050] [<ffffffff81141628>] ? poll_select_copy_remaining+0xf9/0xf9 [141625.761002] [<ffffffff813f0535>] fib6_rule_lookup+0x37/0x5c [141625.766914] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.773260] [<ffffffff813d008c>] ip6_route_output+0x7a/0x82 [141625.779177] [<ffffffff813c44c8>] ip6_dst_lookup_tail+0x53/0x112 [141625.785437] [<ffffffff813c45c3>] ip6_dst_lookup_flow+0x2a/0x6b [141625.791604] [<ffffffff813ddaab>] rawv6_sendmsg+0x407/0x9b6 [141625.797423] [<ffffffff813d7914>] ? do_ipv6_setsockopt.isra.8+0xd87/0xde2 [141625.804464] [<ffffffff8139d4b4>] inet_sendmsg+0x57/0x8e [141625.810028] [<ffffffff81329ba3>] sock_sendmsg+0x2e/0x3c [141625.815588] [<ffffffff8132be57>] SyS_sendto+0xfe/0x143 [141625.821063] [<ffffffff813dd551>] ? rawv6_setsockopt+0x5e/0x67 [141625.827146] [<ffffffff8132c9f8>] ? sock_common_setsockopt+0xf/0x11 [141625.833660] [<ffffffff8132c08c>] ? SyS_setsockopt+0x81/0xa2 [141625.839565] [<ffffffff8140ac17>] entry_SYSCALL_64_fastpath+0x12/0x6a Fixes: d52d3997f843 ("pv6: Create percpu rt6_info") Signed-off-by: Martin KaFai Lau <kafai@fb.com> CC: Hannes Frederic Sowa <hannes@stressinduktion.org> Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-14 18:05:54 +00:00
dst_hold(&net->ipv6.ip6_null_entry->dst);
return net->ipv6.ip6_null_entry;
}
ipv6: Fix a potential deadlock when creating pcpu rt rt6_make_pcpu_route() is called under read_lock(&table->tb6_lock). rt6_make_pcpu_route() calls ip6_rt_pcpu_alloc(rt) which then calls dst_alloc(). dst_alloc() _may_ call ip6_dst_gc() which takes the write_lock(&tabl->tb6_lock). A visualized version: read_lock(&table->tb6_lock); rt6_make_pcpu_route(); => ip6_rt_pcpu_alloc(); => dst_alloc(); => ip6_dst_gc(); => write_lock(&table->tb6_lock); /* oops */ The fix is to do a read_unlock first before calling ip6_rt_pcpu_alloc(). A reported stack: [141625.537638] INFO: rcu_sched self-detected stall on CPU { 27} (t=60000 jiffies g=4159086 c=4159085 q=2139) [141625.547469] Task dump for CPU 27: [141625.550881] mtr R running task 0 22121 22081 0x00000008 [141625.558069] 0000000000000000 ffff88103f363d98 ffffffff8106e488 000000000000001b [141625.565641] ffffffff81684900 ffff88103f363db8 ffffffff810702b0 0000000008000000 [141625.573220] ffffffff81684900 ffff88103f363de8 ffffffff8108df9f ffff88103f375a00 [141625.580803] Call Trace: [141625.583345] <IRQ> [<ffffffff8106e488>] sched_show_task+0xc1/0xc6 [141625.589650] [<ffffffff810702b0>] dump_cpu_task+0x35/0x39 [141625.595144] [<ffffffff8108df9f>] rcu_dump_cpu_stacks+0x6a/0x8c [141625.601320] [<ffffffff81090606>] rcu_check_callbacks+0x1f6/0x5d4 [141625.607669] [<ffffffff810940c8>] update_process_times+0x2a/0x4f [141625.613925] [<ffffffff8109fbee>] tick_sched_handle+0x32/0x3e [141625.619923] [<ffffffff8109fc2f>] tick_sched_timer+0x35/0x5c [141625.625830] [<ffffffff81094a1f>] __hrtimer_run_queues+0x8f/0x18d [141625.632171] [<ffffffff81094c9e>] hrtimer_interrupt+0xa0/0x166 [141625.638258] [<ffffffff8102bf2a>] local_apic_timer_interrupt+0x4e/0x52 [141625.645036] [<ffffffff8102c36f>] smp_apic_timer_interrupt+0x39/0x4a [141625.651643] [<ffffffff8140b9e8>] apic_timer_interrupt+0x68/0x70 [141625.657895] <EOI> [<ffffffff81346ee8>] ? dst_destroy+0x7c/0xb5 [141625.664188] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.670272] [<ffffffff81082b45>] ? queue_write_lock_slowpath+0x60/0x6f [141625.677140] [<ffffffff8140aa33>] _raw_write_lock_bh+0x23/0x25 [141625.683218] [<ffffffff813d4553>] __fib6_clean_all+0x40/0x82 [141625.689124] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.695207] [<ffffffff813d6058>] fib6_clean_all+0xe/0x10 [141625.700854] [<ffffffff813d60d3>] fib6_run_gc+0x79/0xc8 [141625.706329] [<ffffffff813d0510>] ip6_dst_gc+0x85/0xf9 [141625.711718] [<ffffffff81346d68>] dst_alloc+0x55/0x159 [141625.717105] [<ffffffff813d09b5>] __ip6_dst_alloc.isra.32+0x19/0x63 [141625.723620] [<ffffffff813d1830>] ip6_pol_route+0x36a/0x3e8 [141625.729441] [<ffffffff813d18d6>] ip6_pol_route_output+0x11/0x13 [141625.735700] [<ffffffff813f02c8>] fib6_rule_action+0xa7/0x1bf [141625.741698] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.748043] [<ffffffff81357c48>] fib_rules_lookup+0xb5/0x12a [141625.754050] [<ffffffff81141628>] ? poll_select_copy_remaining+0xf9/0xf9 [141625.761002] [<ffffffff813f0535>] fib6_rule_lookup+0x37/0x5c [141625.766914] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.773260] [<ffffffff813d008c>] ip6_route_output+0x7a/0x82 [141625.779177] [<ffffffff813c44c8>] ip6_dst_lookup_tail+0x53/0x112 [141625.785437] [<ffffffff813c45c3>] ip6_dst_lookup_flow+0x2a/0x6b [141625.791604] [<ffffffff813ddaab>] rawv6_sendmsg+0x407/0x9b6 [141625.797423] [<ffffffff813d7914>] ? do_ipv6_setsockopt.isra.8+0xd87/0xde2 [141625.804464] [<ffffffff8139d4b4>] inet_sendmsg+0x57/0x8e [141625.810028] [<ffffffff81329ba3>] sock_sendmsg+0x2e/0x3c [141625.815588] [<ffffffff8132be57>] SyS_sendto+0xfe/0x143 [141625.821063] [<ffffffff813dd551>] ? rawv6_setsockopt+0x5e/0x67 [141625.827146] [<ffffffff8132c9f8>] ? sock_common_setsockopt+0xf/0x11 [141625.833660] [<ffffffff8132c08c>] ? SyS_setsockopt+0x81/0xa2 [141625.839565] [<ffffffff8140ac17>] entry_SYSCALL_64_fastpath+0x12/0x6a Fixes: d52d3997f843 ("pv6: Create percpu rt6_info") Signed-off-by: Martin KaFai Lau <kafai@fb.com> CC: Hannes Frederic Sowa <hannes@stressinduktion.org> Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-14 18:05:54 +00:00
read_lock_bh(&table->tb6_lock);
if (rt->rt6i_pcpu) {
p = this_cpu_ptr(rt->rt6i_pcpu);
prev = cmpxchg(p, NULL, pcpu_rt);
if (prev) {
/* If someone did it before us, return prev instead */
dst_destroy(&pcpu_rt->dst);
pcpu_rt = prev;
}
} else {
/* rt has been removed from the fib6 tree
* before we have a chance to acquire the read_lock.
* In this case, don't brother to create a pcpu rt
* since rt is going away anyway. The next
* dst_check() will trigger a re-lookup.
*/
dst_destroy(&pcpu_rt->dst);
ipv6: Fix a potential deadlock when creating pcpu rt rt6_make_pcpu_route() is called under read_lock(&table->tb6_lock). rt6_make_pcpu_route() calls ip6_rt_pcpu_alloc(rt) which then calls dst_alloc(). dst_alloc() _may_ call ip6_dst_gc() which takes the write_lock(&tabl->tb6_lock). A visualized version: read_lock(&table->tb6_lock); rt6_make_pcpu_route(); => ip6_rt_pcpu_alloc(); => dst_alloc(); => ip6_dst_gc(); => write_lock(&table->tb6_lock); /* oops */ The fix is to do a read_unlock first before calling ip6_rt_pcpu_alloc(). A reported stack: [141625.537638] INFO: rcu_sched self-detected stall on CPU { 27} (t=60000 jiffies g=4159086 c=4159085 q=2139) [141625.547469] Task dump for CPU 27: [141625.550881] mtr R running task 0 22121 22081 0x00000008 [141625.558069] 0000000000000000 ffff88103f363d98 ffffffff8106e488 000000000000001b [141625.565641] ffffffff81684900 ffff88103f363db8 ffffffff810702b0 0000000008000000 [141625.573220] ffffffff81684900 ffff88103f363de8 ffffffff8108df9f ffff88103f375a00 [141625.580803] Call Trace: [141625.583345] <IRQ> [<ffffffff8106e488>] sched_show_task+0xc1/0xc6 [141625.589650] [<ffffffff810702b0>] dump_cpu_task+0x35/0x39 [141625.595144] [<ffffffff8108df9f>] rcu_dump_cpu_stacks+0x6a/0x8c [141625.601320] [<ffffffff81090606>] rcu_check_callbacks+0x1f6/0x5d4 [141625.607669] [<ffffffff810940c8>] update_process_times+0x2a/0x4f [141625.613925] [<ffffffff8109fbee>] tick_sched_handle+0x32/0x3e [141625.619923] [<ffffffff8109fc2f>] tick_sched_timer+0x35/0x5c [141625.625830] [<ffffffff81094a1f>] __hrtimer_run_queues+0x8f/0x18d [141625.632171] [<ffffffff81094c9e>] hrtimer_interrupt+0xa0/0x166 [141625.638258] [<ffffffff8102bf2a>] local_apic_timer_interrupt+0x4e/0x52 [141625.645036] [<ffffffff8102c36f>] smp_apic_timer_interrupt+0x39/0x4a [141625.651643] [<ffffffff8140b9e8>] apic_timer_interrupt+0x68/0x70 [141625.657895] <EOI> [<ffffffff81346ee8>] ? dst_destroy+0x7c/0xb5 [141625.664188] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.670272] [<ffffffff81082b45>] ? queue_write_lock_slowpath+0x60/0x6f [141625.677140] [<ffffffff8140aa33>] _raw_write_lock_bh+0x23/0x25 [141625.683218] [<ffffffff813d4553>] __fib6_clean_all+0x40/0x82 [141625.689124] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.695207] [<ffffffff813d6058>] fib6_clean_all+0xe/0x10 [141625.700854] [<ffffffff813d60d3>] fib6_run_gc+0x79/0xc8 [141625.706329] [<ffffffff813d0510>] ip6_dst_gc+0x85/0xf9 [141625.711718] [<ffffffff81346d68>] dst_alloc+0x55/0x159 [141625.717105] [<ffffffff813d09b5>] __ip6_dst_alloc.isra.32+0x19/0x63 [141625.723620] [<ffffffff813d1830>] ip6_pol_route+0x36a/0x3e8 [141625.729441] [<ffffffff813d18d6>] ip6_pol_route_output+0x11/0x13 [141625.735700] [<ffffffff813f02c8>] fib6_rule_action+0xa7/0x1bf [141625.741698] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.748043] [<ffffffff81357c48>] fib_rules_lookup+0xb5/0x12a [141625.754050] [<ffffffff81141628>] ? poll_select_copy_remaining+0xf9/0xf9 [141625.761002] [<ffffffff813f0535>] fib6_rule_lookup+0x37/0x5c [141625.766914] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.773260] [<ffffffff813d008c>] ip6_route_output+0x7a/0x82 [141625.779177] [<ffffffff813c44c8>] ip6_dst_lookup_tail+0x53/0x112 [141625.785437] [<ffffffff813c45c3>] ip6_dst_lookup_flow+0x2a/0x6b [141625.791604] [<ffffffff813ddaab>] rawv6_sendmsg+0x407/0x9b6 [141625.797423] [<ffffffff813d7914>] ? do_ipv6_setsockopt.isra.8+0xd87/0xde2 [141625.804464] [<ffffffff8139d4b4>] inet_sendmsg+0x57/0x8e [141625.810028] [<ffffffff81329ba3>] sock_sendmsg+0x2e/0x3c [141625.815588] [<ffffffff8132be57>] SyS_sendto+0xfe/0x143 [141625.821063] [<ffffffff813dd551>] ? rawv6_setsockopt+0x5e/0x67 [141625.827146] [<ffffffff8132c9f8>] ? sock_common_setsockopt+0xf/0x11 [141625.833660] [<ffffffff8132c08c>] ? SyS_setsockopt+0x81/0xa2 [141625.839565] [<ffffffff8140ac17>] entry_SYSCALL_64_fastpath+0x12/0x6a Fixes: d52d3997f843 ("pv6: Create percpu rt6_info") Signed-off-by: Martin KaFai Lau <kafai@fb.com> CC: Hannes Frederic Sowa <hannes@stressinduktion.org> Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-14 18:05:54 +00:00
pcpu_rt = rt;
}
dst_hold(&pcpu_rt->dst);
rt6_dst_from_metrics_check(pcpu_rt);
ipv6: Fix a potential deadlock when creating pcpu rt rt6_make_pcpu_route() is called under read_lock(&table->tb6_lock). rt6_make_pcpu_route() calls ip6_rt_pcpu_alloc(rt) which then calls dst_alloc(). dst_alloc() _may_ call ip6_dst_gc() which takes the write_lock(&tabl->tb6_lock). A visualized version: read_lock(&table->tb6_lock); rt6_make_pcpu_route(); => ip6_rt_pcpu_alloc(); => dst_alloc(); => ip6_dst_gc(); => write_lock(&table->tb6_lock); /* oops */ The fix is to do a read_unlock first before calling ip6_rt_pcpu_alloc(). A reported stack: [141625.537638] INFO: rcu_sched self-detected stall on CPU { 27} (t=60000 jiffies g=4159086 c=4159085 q=2139) [141625.547469] Task dump for CPU 27: [141625.550881] mtr R running task 0 22121 22081 0x00000008 [141625.558069] 0000000000000000 ffff88103f363d98 ffffffff8106e488 000000000000001b [141625.565641] ffffffff81684900 ffff88103f363db8 ffffffff810702b0 0000000008000000 [141625.573220] ffffffff81684900 ffff88103f363de8 ffffffff8108df9f ffff88103f375a00 [141625.580803] Call Trace: [141625.583345] <IRQ> [<ffffffff8106e488>] sched_show_task+0xc1/0xc6 [141625.589650] [<ffffffff810702b0>] dump_cpu_task+0x35/0x39 [141625.595144] [<ffffffff8108df9f>] rcu_dump_cpu_stacks+0x6a/0x8c [141625.601320] [<ffffffff81090606>] rcu_check_callbacks+0x1f6/0x5d4 [141625.607669] [<ffffffff810940c8>] update_process_times+0x2a/0x4f [141625.613925] [<ffffffff8109fbee>] tick_sched_handle+0x32/0x3e [141625.619923] [<ffffffff8109fc2f>] tick_sched_timer+0x35/0x5c [141625.625830] [<ffffffff81094a1f>] __hrtimer_run_queues+0x8f/0x18d [141625.632171] [<ffffffff81094c9e>] hrtimer_interrupt+0xa0/0x166 [141625.638258] [<ffffffff8102bf2a>] local_apic_timer_interrupt+0x4e/0x52 [141625.645036] [<ffffffff8102c36f>] smp_apic_timer_interrupt+0x39/0x4a [141625.651643] [<ffffffff8140b9e8>] apic_timer_interrupt+0x68/0x70 [141625.657895] <EOI> [<ffffffff81346ee8>] ? dst_destroy+0x7c/0xb5 [141625.664188] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.670272] [<ffffffff81082b45>] ? queue_write_lock_slowpath+0x60/0x6f [141625.677140] [<ffffffff8140aa33>] _raw_write_lock_bh+0x23/0x25 [141625.683218] [<ffffffff813d4553>] __fib6_clean_all+0x40/0x82 [141625.689124] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.695207] [<ffffffff813d6058>] fib6_clean_all+0xe/0x10 [141625.700854] [<ffffffff813d60d3>] fib6_run_gc+0x79/0xc8 [141625.706329] [<ffffffff813d0510>] ip6_dst_gc+0x85/0xf9 [141625.711718] [<ffffffff81346d68>] dst_alloc+0x55/0x159 [141625.717105] [<ffffffff813d09b5>] __ip6_dst_alloc.isra.32+0x19/0x63 [141625.723620] [<ffffffff813d1830>] ip6_pol_route+0x36a/0x3e8 [141625.729441] [<ffffffff813d18d6>] ip6_pol_route_output+0x11/0x13 [141625.735700] [<ffffffff813f02c8>] fib6_rule_action+0xa7/0x1bf [141625.741698] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.748043] [<ffffffff81357c48>] fib_rules_lookup+0xb5/0x12a [141625.754050] [<ffffffff81141628>] ? poll_select_copy_remaining+0xf9/0xf9 [141625.761002] [<ffffffff813f0535>] fib6_rule_lookup+0x37/0x5c [141625.766914] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.773260] [<ffffffff813d008c>] ip6_route_output+0x7a/0x82 [141625.779177] [<ffffffff813c44c8>] ip6_dst_lookup_tail+0x53/0x112 [141625.785437] [<ffffffff813c45c3>] ip6_dst_lookup_flow+0x2a/0x6b [141625.791604] [<ffffffff813ddaab>] rawv6_sendmsg+0x407/0x9b6 [141625.797423] [<ffffffff813d7914>] ? do_ipv6_setsockopt.isra.8+0xd87/0xde2 [141625.804464] [<ffffffff8139d4b4>] inet_sendmsg+0x57/0x8e [141625.810028] [<ffffffff81329ba3>] sock_sendmsg+0x2e/0x3c [141625.815588] [<ffffffff8132be57>] SyS_sendto+0xfe/0x143 [141625.821063] [<ffffffff813dd551>] ? rawv6_setsockopt+0x5e/0x67 [141625.827146] [<ffffffff8132c9f8>] ? sock_common_setsockopt+0xf/0x11 [141625.833660] [<ffffffff8132c08c>] ? SyS_setsockopt+0x81/0xa2 [141625.839565] [<ffffffff8140ac17>] entry_SYSCALL_64_fastpath+0x12/0x6a Fixes: d52d3997f843 ("pv6: Create percpu rt6_info") Signed-off-by: Martin KaFai Lau <kafai@fb.com> CC: Hannes Frederic Sowa <hannes@stressinduktion.org> Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-14 18:05:54 +00:00
read_unlock_bh(&table->tb6_lock);
return pcpu_rt;
}
static struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, int oif,
struct flowi6 *fl6, int flags)
{
struct fib6_node *fn, *saved_fn;
struct rt6_info *rt;
int strict = 0;
strict |= flags & RT6_LOOKUP_F_IFACE;
if (net->ipv6.devconf_all->forwarding == 0)
strict |= RT6_LOOKUP_F_REACHABLE;
read_lock_bh(&table->tb6_lock);
fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr);
saved_fn = fn;
if (fl6->flowi6_flags & FLOWI_FLAG_SKIP_NH_OIF)
oif = 0;
redo_rt6_select:
rt = rt6_select(fn, oif, strict);
if (rt->rt6i_nsiblings)
rt = rt6_multipath_select(rt, fl6, oif, strict);
if (rt == net->ipv6.ip6_null_entry) {
fn = fib6_backtrack(fn, &fl6->saddr);
if (fn)
goto redo_rt6_select;
else if (strict & RT6_LOOKUP_F_REACHABLE) {
/* also consider unreachable route */
strict &= ~RT6_LOOKUP_F_REACHABLE;
fn = saved_fn;
goto redo_rt6_select;
}
}
if (rt == net->ipv6.ip6_null_entry || (rt->rt6i_flags & RTF_CACHE)) {
dst_use(&rt->dst, jiffies);
read_unlock_bh(&table->tb6_lock);
rt6_dst_from_metrics_check(rt);
return rt;
} else if (unlikely((fl6->flowi6_flags & FLOWI_FLAG_KNOWN_NH) &&
!(rt->rt6i_flags & RTF_GATEWAY))) {
/* Create a RTF_CACHE clone which will not be
* owned by the fib6 tree. It is for the special case where
* the daddr in the skb during the neighbor look-up is different
* from the fl6->daddr used to look-up route here.
*/
struct rt6_info *uncached_rt;
dst_use(&rt->dst, jiffies);
read_unlock_bh(&table->tb6_lock);
uncached_rt = ip6_rt_cache_alloc(rt, &fl6->daddr, NULL);
dst_release(&rt->dst);
if (uncached_rt)
rt6_uncached_list_add(uncached_rt);
else
uncached_rt = net->ipv6.ip6_null_entry;
dst_hold(&uncached_rt->dst);
return uncached_rt;
} else {
/* Get a percpu copy */
struct rt6_info *pcpu_rt;
rt->dst.lastuse = jiffies;
rt->dst.__use++;
pcpu_rt = rt6_get_pcpu_route(rt);
ipv6: Fix a potential deadlock when creating pcpu rt rt6_make_pcpu_route() is called under read_lock(&table->tb6_lock). rt6_make_pcpu_route() calls ip6_rt_pcpu_alloc(rt) which then calls dst_alloc(). dst_alloc() _may_ call ip6_dst_gc() which takes the write_lock(&tabl->tb6_lock). A visualized version: read_lock(&table->tb6_lock); rt6_make_pcpu_route(); => ip6_rt_pcpu_alloc(); => dst_alloc(); => ip6_dst_gc(); => write_lock(&table->tb6_lock); /* oops */ The fix is to do a read_unlock first before calling ip6_rt_pcpu_alloc(). A reported stack: [141625.537638] INFO: rcu_sched self-detected stall on CPU { 27} (t=60000 jiffies g=4159086 c=4159085 q=2139) [141625.547469] Task dump for CPU 27: [141625.550881] mtr R running task 0 22121 22081 0x00000008 [141625.558069] 0000000000000000 ffff88103f363d98 ffffffff8106e488 000000000000001b [141625.565641] ffffffff81684900 ffff88103f363db8 ffffffff810702b0 0000000008000000 [141625.573220] ffffffff81684900 ffff88103f363de8 ffffffff8108df9f ffff88103f375a00 [141625.580803] Call Trace: [141625.583345] <IRQ> [<ffffffff8106e488>] sched_show_task+0xc1/0xc6 [141625.589650] [<ffffffff810702b0>] dump_cpu_task+0x35/0x39 [141625.595144] [<ffffffff8108df9f>] rcu_dump_cpu_stacks+0x6a/0x8c [141625.601320] [<ffffffff81090606>] rcu_check_callbacks+0x1f6/0x5d4 [141625.607669] [<ffffffff810940c8>] update_process_times+0x2a/0x4f [141625.613925] [<ffffffff8109fbee>] tick_sched_handle+0x32/0x3e [141625.619923] [<ffffffff8109fc2f>] tick_sched_timer+0x35/0x5c [141625.625830] [<ffffffff81094a1f>] __hrtimer_run_queues+0x8f/0x18d [141625.632171] [<ffffffff81094c9e>] hrtimer_interrupt+0xa0/0x166 [141625.638258] [<ffffffff8102bf2a>] local_apic_timer_interrupt+0x4e/0x52 [141625.645036] [<ffffffff8102c36f>] smp_apic_timer_interrupt+0x39/0x4a [141625.651643] [<ffffffff8140b9e8>] apic_timer_interrupt+0x68/0x70 [141625.657895] <EOI> [<ffffffff81346ee8>] ? dst_destroy+0x7c/0xb5 [141625.664188] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.670272] [<ffffffff81082b45>] ? queue_write_lock_slowpath+0x60/0x6f [141625.677140] [<ffffffff8140aa33>] _raw_write_lock_bh+0x23/0x25 [141625.683218] [<ffffffff813d4553>] __fib6_clean_all+0x40/0x82 [141625.689124] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.695207] [<ffffffff813d6058>] fib6_clean_all+0xe/0x10 [141625.700854] [<ffffffff813d60d3>] fib6_run_gc+0x79/0xc8 [141625.706329] [<ffffffff813d0510>] ip6_dst_gc+0x85/0xf9 [141625.711718] [<ffffffff81346d68>] dst_alloc+0x55/0x159 [141625.717105] [<ffffffff813d09b5>] __ip6_dst_alloc.isra.32+0x19/0x63 [141625.723620] [<ffffffff813d1830>] ip6_pol_route+0x36a/0x3e8 [141625.729441] [<ffffffff813d18d6>] ip6_pol_route_output+0x11/0x13 [141625.735700] [<ffffffff813f02c8>] fib6_rule_action+0xa7/0x1bf [141625.741698] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.748043] [<ffffffff81357c48>] fib_rules_lookup+0xb5/0x12a [141625.754050] [<ffffffff81141628>] ? poll_select_copy_remaining+0xf9/0xf9 [141625.761002] [<ffffffff813f0535>] fib6_rule_lookup+0x37/0x5c [141625.766914] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.773260] [<ffffffff813d008c>] ip6_route_output+0x7a/0x82 [141625.779177] [<ffffffff813c44c8>] ip6_dst_lookup_tail+0x53/0x112 [141625.785437] [<ffffffff813c45c3>] ip6_dst_lookup_flow+0x2a/0x6b [141625.791604] [<ffffffff813ddaab>] rawv6_sendmsg+0x407/0x9b6 [141625.797423] [<ffffffff813d7914>] ? do_ipv6_setsockopt.isra.8+0xd87/0xde2 [141625.804464] [<ffffffff8139d4b4>] inet_sendmsg+0x57/0x8e [141625.810028] [<ffffffff81329ba3>] sock_sendmsg+0x2e/0x3c [141625.815588] [<ffffffff8132be57>] SyS_sendto+0xfe/0x143 [141625.821063] [<ffffffff813dd551>] ? rawv6_setsockopt+0x5e/0x67 [141625.827146] [<ffffffff8132c9f8>] ? sock_common_setsockopt+0xf/0x11 [141625.833660] [<ffffffff8132c08c>] ? SyS_setsockopt+0x81/0xa2 [141625.839565] [<ffffffff8140ac17>] entry_SYSCALL_64_fastpath+0x12/0x6a Fixes: d52d3997f843 ("pv6: Create percpu rt6_info") Signed-off-by: Martin KaFai Lau <kafai@fb.com> CC: Hannes Frederic Sowa <hannes@stressinduktion.org> Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-14 18:05:54 +00:00
if (pcpu_rt) {
read_unlock_bh(&table->tb6_lock);
} else {
/* We have to do the read_unlock first
* because rt6_make_pcpu_route() may trigger
* ip6_dst_gc() which will take the write_lock.
*/
dst_hold(&rt->dst);
read_unlock_bh(&table->tb6_lock);
pcpu_rt = rt6_make_pcpu_route(rt);
ipv6: Fix a potential deadlock when creating pcpu rt rt6_make_pcpu_route() is called under read_lock(&table->tb6_lock). rt6_make_pcpu_route() calls ip6_rt_pcpu_alloc(rt) which then calls dst_alloc(). dst_alloc() _may_ call ip6_dst_gc() which takes the write_lock(&tabl->tb6_lock). A visualized version: read_lock(&table->tb6_lock); rt6_make_pcpu_route(); => ip6_rt_pcpu_alloc(); => dst_alloc(); => ip6_dst_gc(); => write_lock(&table->tb6_lock); /* oops */ The fix is to do a read_unlock first before calling ip6_rt_pcpu_alloc(). A reported stack: [141625.537638] INFO: rcu_sched self-detected stall on CPU { 27} (t=60000 jiffies g=4159086 c=4159085 q=2139) [141625.547469] Task dump for CPU 27: [141625.550881] mtr R running task 0 22121 22081 0x00000008 [141625.558069] 0000000000000000 ffff88103f363d98 ffffffff8106e488 000000000000001b [141625.565641] ffffffff81684900 ffff88103f363db8 ffffffff810702b0 0000000008000000 [141625.573220] ffffffff81684900 ffff88103f363de8 ffffffff8108df9f ffff88103f375a00 [141625.580803] Call Trace: [141625.583345] <IRQ> [<ffffffff8106e488>] sched_show_task+0xc1/0xc6 [141625.589650] [<ffffffff810702b0>] dump_cpu_task+0x35/0x39 [141625.595144] [<ffffffff8108df9f>] rcu_dump_cpu_stacks+0x6a/0x8c [141625.601320] [<ffffffff81090606>] rcu_check_callbacks+0x1f6/0x5d4 [141625.607669] [<ffffffff810940c8>] update_process_times+0x2a/0x4f [141625.613925] [<ffffffff8109fbee>] tick_sched_handle+0x32/0x3e [141625.619923] [<ffffffff8109fc2f>] tick_sched_timer+0x35/0x5c [141625.625830] [<ffffffff81094a1f>] __hrtimer_run_queues+0x8f/0x18d [141625.632171] [<ffffffff81094c9e>] hrtimer_interrupt+0xa0/0x166 [141625.638258] [<ffffffff8102bf2a>] local_apic_timer_interrupt+0x4e/0x52 [141625.645036] [<ffffffff8102c36f>] smp_apic_timer_interrupt+0x39/0x4a [141625.651643] [<ffffffff8140b9e8>] apic_timer_interrupt+0x68/0x70 [141625.657895] <EOI> [<ffffffff81346ee8>] ? dst_destroy+0x7c/0xb5 [141625.664188] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.670272] [<ffffffff81082b45>] ? queue_write_lock_slowpath+0x60/0x6f [141625.677140] [<ffffffff8140aa33>] _raw_write_lock_bh+0x23/0x25 [141625.683218] [<ffffffff813d4553>] __fib6_clean_all+0x40/0x82 [141625.689124] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.695207] [<ffffffff813d6058>] fib6_clean_all+0xe/0x10 [141625.700854] [<ffffffff813d60d3>] fib6_run_gc+0x79/0xc8 [141625.706329] [<ffffffff813d0510>] ip6_dst_gc+0x85/0xf9 [141625.711718] [<ffffffff81346d68>] dst_alloc+0x55/0x159 [141625.717105] [<ffffffff813d09b5>] __ip6_dst_alloc.isra.32+0x19/0x63 [141625.723620] [<ffffffff813d1830>] ip6_pol_route+0x36a/0x3e8 [141625.729441] [<ffffffff813d18d6>] ip6_pol_route_output+0x11/0x13 [141625.735700] [<ffffffff813f02c8>] fib6_rule_action+0xa7/0x1bf [141625.741698] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.748043] [<ffffffff81357c48>] fib_rules_lookup+0xb5/0x12a [141625.754050] [<ffffffff81141628>] ? poll_select_copy_remaining+0xf9/0xf9 [141625.761002] [<ffffffff813f0535>] fib6_rule_lookup+0x37/0x5c [141625.766914] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.773260] [<ffffffff813d008c>] ip6_route_output+0x7a/0x82 [141625.779177] [<ffffffff813c44c8>] ip6_dst_lookup_tail+0x53/0x112 [141625.785437] [<ffffffff813c45c3>] ip6_dst_lookup_flow+0x2a/0x6b [141625.791604] [<ffffffff813ddaab>] rawv6_sendmsg+0x407/0x9b6 [141625.797423] [<ffffffff813d7914>] ? do_ipv6_setsockopt.isra.8+0xd87/0xde2 [141625.804464] [<ffffffff8139d4b4>] inet_sendmsg+0x57/0x8e [141625.810028] [<ffffffff81329ba3>] sock_sendmsg+0x2e/0x3c [141625.815588] [<ffffffff8132be57>] SyS_sendto+0xfe/0x143 [141625.821063] [<ffffffff813dd551>] ? rawv6_setsockopt+0x5e/0x67 [141625.827146] [<ffffffff8132c9f8>] ? sock_common_setsockopt+0xf/0x11 [141625.833660] [<ffffffff8132c08c>] ? SyS_setsockopt+0x81/0xa2 [141625.839565] [<ffffffff8140ac17>] entry_SYSCALL_64_fastpath+0x12/0x6a Fixes: d52d3997f843 ("pv6: Create percpu rt6_info") Signed-off-by: Martin KaFai Lau <kafai@fb.com> CC: Hannes Frederic Sowa <hannes@stressinduktion.org> Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-14 18:05:54 +00:00
dst_release(&rt->dst);
}
return pcpu_rt;
ipv6: Fix a potential deadlock when creating pcpu rt rt6_make_pcpu_route() is called under read_lock(&table->tb6_lock). rt6_make_pcpu_route() calls ip6_rt_pcpu_alloc(rt) which then calls dst_alloc(). dst_alloc() _may_ call ip6_dst_gc() which takes the write_lock(&tabl->tb6_lock). A visualized version: read_lock(&table->tb6_lock); rt6_make_pcpu_route(); => ip6_rt_pcpu_alloc(); => dst_alloc(); => ip6_dst_gc(); => write_lock(&table->tb6_lock); /* oops */ The fix is to do a read_unlock first before calling ip6_rt_pcpu_alloc(). A reported stack: [141625.537638] INFO: rcu_sched self-detected stall on CPU { 27} (t=60000 jiffies g=4159086 c=4159085 q=2139) [141625.547469] Task dump for CPU 27: [141625.550881] mtr R running task 0 22121 22081 0x00000008 [141625.558069] 0000000000000000 ffff88103f363d98 ffffffff8106e488 000000000000001b [141625.565641] ffffffff81684900 ffff88103f363db8 ffffffff810702b0 0000000008000000 [141625.573220] ffffffff81684900 ffff88103f363de8 ffffffff8108df9f ffff88103f375a00 [141625.580803] Call Trace: [141625.583345] <IRQ> [<ffffffff8106e488>] sched_show_task+0xc1/0xc6 [141625.589650] [<ffffffff810702b0>] dump_cpu_task+0x35/0x39 [141625.595144] [<ffffffff8108df9f>] rcu_dump_cpu_stacks+0x6a/0x8c [141625.601320] [<ffffffff81090606>] rcu_check_callbacks+0x1f6/0x5d4 [141625.607669] [<ffffffff810940c8>] update_process_times+0x2a/0x4f [141625.613925] [<ffffffff8109fbee>] tick_sched_handle+0x32/0x3e [141625.619923] [<ffffffff8109fc2f>] tick_sched_timer+0x35/0x5c [141625.625830] [<ffffffff81094a1f>] __hrtimer_run_queues+0x8f/0x18d [141625.632171] [<ffffffff81094c9e>] hrtimer_interrupt+0xa0/0x166 [141625.638258] [<ffffffff8102bf2a>] local_apic_timer_interrupt+0x4e/0x52 [141625.645036] [<ffffffff8102c36f>] smp_apic_timer_interrupt+0x39/0x4a [141625.651643] [<ffffffff8140b9e8>] apic_timer_interrupt+0x68/0x70 [141625.657895] <EOI> [<ffffffff81346ee8>] ? dst_destroy+0x7c/0xb5 [141625.664188] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.670272] [<ffffffff81082b45>] ? queue_write_lock_slowpath+0x60/0x6f [141625.677140] [<ffffffff8140aa33>] _raw_write_lock_bh+0x23/0x25 [141625.683218] [<ffffffff813d4553>] __fib6_clean_all+0x40/0x82 [141625.689124] [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20 [141625.695207] [<ffffffff813d6058>] fib6_clean_all+0xe/0x10 [141625.700854] [<ffffffff813d60d3>] fib6_run_gc+0x79/0xc8 [141625.706329] [<ffffffff813d0510>] ip6_dst_gc+0x85/0xf9 [141625.711718] [<ffffffff81346d68>] dst_alloc+0x55/0x159 [141625.717105] [<ffffffff813d09b5>] __ip6_dst_alloc.isra.32+0x19/0x63 [141625.723620] [<ffffffff813d1830>] ip6_pol_route+0x36a/0x3e8 [141625.729441] [<ffffffff813d18d6>] ip6_pol_route_output+0x11/0x13 [141625.735700] [<ffffffff813f02c8>] fib6_rule_action+0xa7/0x1bf [141625.741698] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.748043] [<ffffffff81357c48>] fib_rules_lookup+0xb5/0x12a [141625.754050] [<ffffffff81141628>] ? poll_select_copy_remaining+0xf9/0xf9 [141625.761002] [<ffffffff813f0535>] fib6_rule_lookup+0x37/0x5c [141625.766914] [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17 [141625.773260] [<ffffffff813d008c>] ip6_route_output+0x7a/0x82 [141625.779177] [<ffffffff813c44c8>] ip6_dst_lookup_tail+0x53/0x112 [141625.785437] [<ffffffff813c45c3>] ip6_dst_lookup_flow+0x2a/0x6b [141625.791604] [<ffffffff813ddaab>] rawv6_sendmsg+0x407/0x9b6 [141625.797423] [<ffffffff813d7914>] ? do_ipv6_setsockopt.isra.8+0xd87/0xde2 [141625.804464] [<ffffffff8139d4b4>] inet_sendmsg+0x57/0x8e [141625.810028] [<ffffffff81329ba3>] sock_sendmsg+0x2e/0x3c [141625.815588] [<ffffffff8132be57>] SyS_sendto+0xfe/0x143 [141625.821063] [<ffffffff813dd551>] ? rawv6_setsockopt+0x5e/0x67 [141625.827146] [<ffffffff8132c9f8>] ? sock_common_setsockopt+0xf/0x11 [141625.833660] [<ffffffff8132c08c>] ? SyS_setsockopt+0x81/0xa2 [141625.839565] [<ffffffff8140ac17>] entry_SYSCALL_64_fastpath+0x12/0x6a Fixes: d52d3997f843 ("pv6: Create percpu rt6_info") Signed-off-by: Martin KaFai Lau <kafai@fb.com> CC: Hannes Frederic Sowa <hannes@stressinduktion.org> Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-14 18:05:54 +00:00
}
}
static struct rt6_info *ip6_pol_route_input(struct net *net, struct fib6_table *table,
struct flowi6 *fl6, int flags)
{
return ip6_pol_route(net, table, fl6->flowi6_iif, fl6, flags);
}
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);
}
void ip6_route_input(struct sk_buff *skb)
{
const struct ipv6hdr *iph = ipv6_hdr(skb);
struct net *net = dev_net(skb->dev);
int flags = RT6_LOOKUP_F_HAS_SADDR;
struct ip_tunnel_info *tun_info;
struct flowi6 fl6 = {
.flowi6_iif = skb->dev->ifindex,
.daddr = iph->daddr,
.saddr = iph->saddr,
.flowlabel = ip6_flowinfo(iph),
.flowi6_mark = skb->mark,
.flowi6_proto = iph->nexthdr,
};
tun_info = skb_tunnel_info(skb);
if (tun_info && !(tun_info->mode & IP_TUNNEL_INFO_TX))
fl6.flowi6_tun_key.tun_id = tun_info->key.tun_id;
skb_dst_drop(skb);
skb_dst_set(skb, ip6_route_input_lookup(net, skb->dev, &fl6, flags));
}
static struct rt6_info *ip6_pol_route_output(struct net *net, struct fib6_table *table,
struct flowi6 *fl6, int flags)
{
return ip6_pol_route(net, table, fl6->flowi6_oif, fl6, flags);
}
struct dst_entry *ip6_route_output(struct net *net, const struct sock *sk,
struct flowi6 *fl6)
{
int flags = 0;
fl6->flowi6_iif = LOOPBACK_IFINDEX;
if ((sk && sk->sk_bound_dev_if) || rt6_need_strict(&fl6->daddr) ||
fl6->flowi6_oif)
flags |= RT6_LOOKUP_F_IFACE;
if (!ipv6_addr_any(&fl6->saddr))
flags |= RT6_LOOKUP_F_HAS_SADDR;
else if (sk)
flags |= rt6_srcprefs2flags(inet6_sk(sk)->srcprefs);
return fib6_rule_lookup(net, fl6, flags, ip6_pol_route_output);
}
EXPORT_SYMBOL(ip6_route_output);
struct dst_entry *ip6_blackhole_route(struct net *net, struct dst_entry *dst_orig)
{
struct rt6_info *rt, *ort = (struct rt6_info *) dst_orig;
struct dst_entry *new = NULL;
rt = dst_alloc(&ip6_dst_blackhole_ops, ort->dst.dev, 1, DST_OBSOLETE_NONE, 0);
if (rt) {
rt6_info_init(rt);
new = &rt->dst;
new->__use = 1;
new->input = dst_discard;
new->output = dst_discard_sk;
dst_copy_metrics(new, &ort->dst);
rt->rt6i_idev = ort->rt6i_idev;
if (rt->rt6i_idev)
in6_dev_hold(rt->rt6i_idev);
rt->rt6i_gateway = ort->rt6i_gateway;
rt->rt6i_flags = ort->rt6i_flags & ~RTF_PCPU;
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);
}
dst_release(dst_orig);
return new ? new : ERR_PTR(-ENOMEM);
}
/*
* Destination cache support functions
*/
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
static void rt6_dst_from_metrics_check(struct rt6_info *rt)
{
if (rt->dst.from &&
dst_metrics_ptr(&rt->dst) != dst_metrics_ptr(rt->dst.from))
dst_init_metrics(&rt->dst, dst_metrics_ptr(rt->dst.from), true);
}
static struct dst_entry *rt6_check(struct rt6_info *rt, u32 cookie)
{
if (!rt->rt6i_node || (rt->rt6i_node->fn_sernum != cookie))
return NULL;
if (rt6_check_expired(rt))
return NULL;
return &rt->dst;
}
static struct dst_entry *rt6_dst_from_check(struct rt6_info *rt, u32 cookie)
{
if (rt->dst.obsolete == DST_OBSOLETE_FORCE_CHK &&
rt6_check((struct rt6_info *)(rt->dst.from), cookie))
return &rt->dst;
else
return NULL;
}
static struct dst_entry *ip6_dst_check(struct dst_entry *dst, u32 cookie)
{
struct rt6_info *rt;
rt = (struct rt6_info *) dst;
/* All IPV6 dsts are created with ->obsolete set to the value
* DST_OBSOLETE_FORCE_CHK which forces validation calls down
* into this function always.
*/
ipv6: ip6_dst_check needs to check for expired dst_entries On receiving a packet too big icmp error we check if our current cached dst_entry in the socket is still valid. This validation check did not care about the expiration of the (cached) route. The error path I traced down: The socket receives a packet too big mtu notification. It still has a valid dst_entry and thus issues the ip6_rt_pmtu_update on this dst_entry, setting RTF_EXPIRE and updates the dst.expiration value (which could fail because of not up-to-date expiration values, see previous patch). In some seldom cases we race with a) the ip6_fib gc or b) another routing lookup which would result in a recreation of the cached rt6_info from its parent non-cached rt6_info. While copying the rt6_info we reinitialize the metrics store by copying it over from the parent thus invalidating the just installed pmtu update (both dsts use the same key to the inetpeer storage). The dst_entry with the just invalidated metrics data would just get its RTF_EXPIRES flag cleared and would continue to stay valid for the socket. We should have not issued the pmtu update on the already expired dst_entry in the first placed. By checking the expiration on the dst entry and doing a relookup in case it is out of date we close the race because we would install a new rt6_info into the fib before we issue the pmtu update, thus closing this race. Not reliably updating the dst.expire value was fixed by the patch "ipv6: reset dst.expires value when clearing expire flag". Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Reported-by: Valentijn Sessink <valentyn@blub.net> Cc: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Reviewed-by: Eric Dumazet <edumazet@google.com> Tested-by: Valentijn Sessink <valentyn@blub.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-24 05:48:24 +00:00
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
rt6_dst_from_metrics_check(rt);
if ((rt->rt6i_flags & RTF_PCPU) || unlikely(dst->flags & DST_NOCACHE))
return rt6_dst_from_check(rt, cookie);
else
return rt6_check(rt, cookie);
}
static struct dst_entry *ip6_negative_advice(struct dst_entry *dst)
{
struct rt6_info *rt = (struct rt6_info *) dst;
if (rt) {
if (rt->rt6i_flags & RTF_CACHE) {
if (rt6_check_expired(rt)) {
ip6_del_rt(rt);
dst = NULL;
}
} else {
dst_release(dst);
dst = NULL;
}
}
return dst;
}
static void ip6_link_failure(struct sk_buff *skb)
{
struct rt6_info *rt;
icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0);
rt = (struct rt6_info *) skb_dst(skb);
if (rt) {
ipv6: in case of link failure remove route directly instead of letting it expire We could end up expiring a route which is part of an ecmp route set. Doing so would invalidate the rt->rt6i_nsiblings calculations and could provoke the following panic: [ 80.144667] ------------[ cut here ]------------ [ 80.145172] kernel BUG at net/ipv6/ip6_fib.c:733! [ 80.145172] invalid opcode: 0000 [#1] SMP [ 80.145172] Modules linked in: 8021q nf_conntrack_netbios_ns nf_conntrack_broadcast ipt_MASQUERADE ip6table_mangle ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 iptable_nat nf_nat_ipv4 nf_nat iptable_mangle nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ebtable_filter ebtables ip6table_filter ip6_tables +snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_page_alloc snd_timer virtio_balloon snd soundcore i2c_piix4 i2c_core virtio_net virtio_blk [ 80.145172] CPU: 1 PID: 786 Comm: ping6 Not tainted 3.10.0+ #118 [ 80.145172] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 80.145172] task: ffff880117fa0000 ti: ffff880118770000 task.ti: ffff880118770000 [ 80.145172] RIP: 0010:[<ffffffff815f3b5d>] [<ffffffff815f3b5d>] fib6_add+0x75d/0x830 [ 80.145172] RSP: 0018:ffff880118771798 EFLAGS: 00010202 [ 80.145172] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff88011350e480 [ 80.145172] RDX: ffff88011350e238 RSI: 0000000000000004 RDI: ffff88011350f738 [ 80.145172] RBP: ffff880118771848 R08: ffff880117903280 R09: 0000000000000001 [ 80.145172] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88011350f680 [ 80.145172] R13: ffff880117903280 R14: ffff880118771890 R15: ffff88011350ef90 [ 80.145172] FS: 00007f02b5127740(0000) GS:ffff88011fd00000(0000) knlGS:0000000000000000 [ 80.145172] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [ 80.145172] CR2: 00007f981322a000 CR3: 00000001181b1000 CR4: 00000000000006e0 [ 80.145172] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 80.145172] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 80.145172] Stack: [ 80.145172] 0000000000000001 ffff880100000000 ffff880100000000 ffff880117903280 [ 80.145172] 0000000000000000 ffff880119a4cf00 0000000000000400 00000000000007fa [ 80.145172] 0000000000000000 0000000000000000 0000000000000000 ffff88011350f680 [ 80.145172] Call Trace: [ 80.145172] [<ffffffff815eeceb>] ? rt6_bind_peer+0x4b/0x90 [ 80.145172] [<ffffffff815ed985>] __ip6_ins_rt+0x45/0x70 [ 80.145172] [<ffffffff815eee35>] ip6_ins_rt+0x35/0x40 [ 80.145172] [<ffffffff815ef1e4>] ip6_pol_route.isra.44+0x3a4/0x4b0 [ 80.145172] [<ffffffff815ef34a>] ip6_pol_route_output+0x2a/0x30 [ 80.145172] [<ffffffff81616077>] fib6_rule_action+0xd7/0x210 [ 80.145172] [<ffffffff815ef320>] ? ip6_pol_route_input+0x30/0x30 [ 80.145172] [<ffffffff81553026>] fib_rules_lookup+0xc6/0x140 [ 80.145172] [<ffffffff81616374>] fib6_rule_lookup+0x44/0x80 [ 80.145172] [<ffffffff815ef320>] ? ip6_pol_route_input+0x30/0x30 [ 80.145172] [<ffffffff815edea3>] ip6_route_output+0x73/0xb0 [ 80.145172] [<ffffffff815dfdf3>] ip6_dst_lookup_tail+0x2c3/0x2e0 [ 80.145172] [<ffffffff813007b1>] ? list_del+0x11/0x40 [ 80.145172] [<ffffffff81082a4c>] ? remove_wait_queue+0x3c/0x50 [ 80.145172] [<ffffffff815dfe4d>] ip6_dst_lookup_flow+0x3d/0xa0 [ 80.145172] [<ffffffff815fda77>] rawv6_sendmsg+0x267/0xc20 [ 80.145172] [<ffffffff815a8a83>] inet_sendmsg+0x63/0xb0 [ 80.145172] [<ffffffff8128eb93>] ? selinux_socket_sendmsg+0x23/0x30 [ 80.145172] [<ffffffff815218d6>] sock_sendmsg+0xa6/0xd0 [ 80.145172] [<ffffffff81524a68>] SYSC_sendto+0x128/0x180 [ 80.145172] [<ffffffff8109825c>] ? update_curr+0xec/0x170 [ 80.145172] [<ffffffff81041d09>] ? kvm_clock_get_cycles+0x9/0x10 [ 80.145172] [<ffffffff810afd1e>] ? __getnstimeofday+0x3e/0xd0 [ 80.145172] [<ffffffff8152509e>] SyS_sendto+0xe/0x10 [ 80.145172] [<ffffffff8164efd9>] system_call_fastpath+0x16/0x1b [ 80.145172] Code: fe ff ff 41 f6 45 2a 06 0f 85 ca fe ff ff 49 8b 7e 08 4c 89 ee e8 94 ef ff ff e9 b9 fe ff ff 48 8b 82 28 05 00 00 e9 01 ff ff ff <0f> 0b 49 8b 54 24 30 0d 00 00 40 00 89 83 14 01 00 00 48 89 53 [ 80.145172] RIP [<ffffffff815f3b5d>] fib6_add+0x75d/0x830 [ 80.145172] RSP <ffff880118771798> [ 80.387413] ---[ end trace 02f20b7a8b81ed95 ]--- [ 80.390154] Kernel panic - not syncing: Fatal exception in interrupt Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com> Cc: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-10 21:00:57 +00:00
if (rt->rt6i_flags & RTF_CACHE) {
dst_hold(&rt->dst);
ip6_del_rt(rt);
ipv6: in case of link failure remove route directly instead of letting it expire We could end up expiring a route which is part of an ecmp route set. Doing so would invalidate the rt->rt6i_nsiblings calculations and could provoke the following panic: [ 80.144667] ------------[ cut here ]------------ [ 80.145172] kernel BUG at net/ipv6/ip6_fib.c:733! [ 80.145172] invalid opcode: 0000 [#1] SMP [ 80.145172] Modules linked in: 8021q nf_conntrack_netbios_ns nf_conntrack_broadcast ipt_MASQUERADE ip6table_mangle ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 iptable_nat nf_nat_ipv4 nf_nat iptable_mangle nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ebtable_filter ebtables ip6table_filter ip6_tables +snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_page_alloc snd_timer virtio_balloon snd soundcore i2c_piix4 i2c_core virtio_net virtio_blk [ 80.145172] CPU: 1 PID: 786 Comm: ping6 Not tainted 3.10.0+ #118 [ 80.145172] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 80.145172] task: ffff880117fa0000 ti: ffff880118770000 task.ti: ffff880118770000 [ 80.145172] RIP: 0010:[<ffffffff815f3b5d>] [<ffffffff815f3b5d>] fib6_add+0x75d/0x830 [ 80.145172] RSP: 0018:ffff880118771798 EFLAGS: 00010202 [ 80.145172] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff88011350e480 [ 80.145172] RDX: ffff88011350e238 RSI: 0000000000000004 RDI: ffff88011350f738 [ 80.145172] RBP: ffff880118771848 R08: ffff880117903280 R09: 0000000000000001 [ 80.145172] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88011350f680 [ 80.145172] R13: ffff880117903280 R14: ffff880118771890 R15: ffff88011350ef90 [ 80.145172] FS: 00007f02b5127740(0000) GS:ffff88011fd00000(0000) knlGS:0000000000000000 [ 80.145172] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [ 80.145172] CR2: 00007f981322a000 CR3: 00000001181b1000 CR4: 00000000000006e0 [ 80.145172] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 80.145172] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 80.145172] Stack: [ 80.145172] 0000000000000001 ffff880100000000 ffff880100000000 ffff880117903280 [ 80.145172] 0000000000000000 ffff880119a4cf00 0000000000000400 00000000000007fa [ 80.145172] 0000000000000000 0000000000000000 0000000000000000 ffff88011350f680 [ 80.145172] Call Trace: [ 80.145172] [<ffffffff815eeceb>] ? rt6_bind_peer+0x4b/0x90 [ 80.145172] [<ffffffff815ed985>] __ip6_ins_rt+0x45/0x70 [ 80.145172] [<ffffffff815eee35>] ip6_ins_rt+0x35/0x40 [ 80.145172] [<ffffffff815ef1e4>] ip6_pol_route.isra.44+0x3a4/0x4b0 [ 80.145172] [<ffffffff815ef34a>] ip6_pol_route_output+0x2a/0x30 [ 80.145172] [<ffffffff81616077>] fib6_rule_action+0xd7/0x210 [ 80.145172] [<ffffffff815ef320>] ? ip6_pol_route_input+0x30/0x30 [ 80.145172] [<ffffffff81553026>] fib_rules_lookup+0xc6/0x140 [ 80.145172] [<ffffffff81616374>] fib6_rule_lookup+0x44/0x80 [ 80.145172] [<ffffffff815ef320>] ? ip6_pol_route_input+0x30/0x30 [ 80.145172] [<ffffffff815edea3>] ip6_route_output+0x73/0xb0 [ 80.145172] [<ffffffff815dfdf3>] ip6_dst_lookup_tail+0x2c3/0x2e0 [ 80.145172] [<ffffffff813007b1>] ? list_del+0x11/0x40 [ 80.145172] [<ffffffff81082a4c>] ? remove_wait_queue+0x3c/0x50 [ 80.145172] [<ffffffff815dfe4d>] ip6_dst_lookup_flow+0x3d/0xa0 [ 80.145172] [<ffffffff815fda77>] rawv6_sendmsg+0x267/0xc20 [ 80.145172] [<ffffffff815a8a83>] inet_sendmsg+0x63/0xb0 [ 80.145172] [<ffffffff8128eb93>] ? selinux_socket_sendmsg+0x23/0x30 [ 80.145172] [<ffffffff815218d6>] sock_sendmsg+0xa6/0xd0 [ 80.145172] [<ffffffff81524a68>] SYSC_sendto+0x128/0x180 [ 80.145172] [<ffffffff8109825c>] ? update_curr+0xec/0x170 [ 80.145172] [<ffffffff81041d09>] ? kvm_clock_get_cycles+0x9/0x10 [ 80.145172] [<ffffffff810afd1e>] ? __getnstimeofday+0x3e/0xd0 [ 80.145172] [<ffffffff8152509e>] SyS_sendto+0xe/0x10 [ 80.145172] [<ffffffff8164efd9>] system_call_fastpath+0x16/0x1b [ 80.145172] Code: fe ff ff 41 f6 45 2a 06 0f 85 ca fe ff ff 49 8b 7e 08 4c 89 ee e8 94 ef ff ff e9 b9 fe ff ff 48 8b 82 28 05 00 00 e9 01 ff ff ff <0f> 0b 49 8b 54 24 30 0d 00 00 40 00 89 83 14 01 00 00 48 89 53 [ 80.145172] RIP [<ffffffff815f3b5d>] fib6_add+0x75d/0x830 [ 80.145172] RSP <ffff880118771798> [ 80.387413] ---[ end trace 02f20b7a8b81ed95 ]--- [ 80.390154] Kernel panic - not syncing: Fatal exception in interrupt Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com> Cc: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-10 21:00:57 +00:00
} else if (rt->rt6i_node && (rt->rt6i_flags & RTF_DEFAULT)) {
rt->rt6i_node->fn_sernum = -1;
ipv6: in case of link failure remove route directly instead of letting it expire We could end up expiring a route which is part of an ecmp route set. Doing so would invalidate the rt->rt6i_nsiblings calculations and could provoke the following panic: [ 80.144667] ------------[ cut here ]------------ [ 80.145172] kernel BUG at net/ipv6/ip6_fib.c:733! [ 80.145172] invalid opcode: 0000 [#1] SMP [ 80.145172] Modules linked in: 8021q nf_conntrack_netbios_ns nf_conntrack_broadcast ipt_MASQUERADE ip6table_mangle ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 iptable_nat nf_nat_ipv4 nf_nat iptable_mangle nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ebtable_filter ebtables ip6table_filter ip6_tables +snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_page_alloc snd_timer virtio_balloon snd soundcore i2c_piix4 i2c_core virtio_net virtio_blk [ 80.145172] CPU: 1 PID: 786 Comm: ping6 Not tainted 3.10.0+ #118 [ 80.145172] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 80.145172] task: ffff880117fa0000 ti: ffff880118770000 task.ti: ffff880118770000 [ 80.145172] RIP: 0010:[<ffffffff815f3b5d>] [<ffffffff815f3b5d>] fib6_add+0x75d/0x830 [ 80.145172] RSP: 0018:ffff880118771798 EFLAGS: 00010202 [ 80.145172] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff88011350e480 [ 80.145172] RDX: ffff88011350e238 RSI: 0000000000000004 RDI: ffff88011350f738 [ 80.145172] RBP: ffff880118771848 R08: ffff880117903280 R09: 0000000000000001 [ 80.145172] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88011350f680 [ 80.145172] R13: ffff880117903280 R14: ffff880118771890 R15: ffff88011350ef90 [ 80.145172] FS: 00007f02b5127740(0000) GS:ffff88011fd00000(0000) knlGS:0000000000000000 [ 80.145172] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [ 80.145172] CR2: 00007f981322a000 CR3: 00000001181b1000 CR4: 00000000000006e0 [ 80.145172] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 80.145172] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 80.145172] Stack: [ 80.145172] 0000000000000001 ffff880100000000 ffff880100000000 ffff880117903280 [ 80.145172] 0000000000000000 ffff880119a4cf00 0000000000000400 00000000000007fa [ 80.145172] 0000000000000000 0000000000000000 0000000000000000 ffff88011350f680 [ 80.145172] Call Trace: [ 80.145172] [<ffffffff815eeceb>] ? rt6_bind_peer+0x4b/0x90 [ 80.145172] [<ffffffff815ed985>] __ip6_ins_rt+0x45/0x70 [ 80.145172] [<ffffffff815eee35>] ip6_ins_rt+0x35/0x40 [ 80.145172] [<ffffffff815ef1e4>] ip6_pol_route.isra.44+0x3a4/0x4b0 [ 80.145172] [<ffffffff815ef34a>] ip6_pol_route_output+0x2a/0x30 [ 80.145172] [<ffffffff81616077>] fib6_rule_action+0xd7/0x210 [ 80.145172] [<ffffffff815ef320>] ? ip6_pol_route_input+0x30/0x30 [ 80.145172] [<ffffffff81553026>] fib_rules_lookup+0xc6/0x140 [ 80.145172] [<ffffffff81616374>] fib6_rule_lookup+0x44/0x80 [ 80.145172] [<ffffffff815ef320>] ? ip6_pol_route_input+0x30/0x30 [ 80.145172] [<ffffffff815edea3>] ip6_route_output+0x73/0xb0 [ 80.145172] [<ffffffff815dfdf3>] ip6_dst_lookup_tail+0x2c3/0x2e0 [ 80.145172] [<ffffffff813007b1>] ? list_del+0x11/0x40 [ 80.145172] [<ffffffff81082a4c>] ? remove_wait_queue+0x3c/0x50 [ 80.145172] [<ffffffff815dfe4d>] ip6_dst_lookup_flow+0x3d/0xa0 [ 80.145172] [<ffffffff815fda77>] rawv6_sendmsg+0x267/0xc20 [ 80.145172] [<ffffffff815a8a83>] inet_sendmsg+0x63/0xb0 [ 80.145172] [<ffffffff8128eb93>] ? selinux_socket_sendmsg+0x23/0x30 [ 80.145172] [<ffffffff815218d6>] sock_sendmsg+0xa6/0xd0 [ 80.145172] [<ffffffff81524a68>] SYSC_sendto+0x128/0x180 [ 80.145172] [<ffffffff8109825c>] ? update_curr+0xec/0x170 [ 80.145172] [<ffffffff81041d09>] ? kvm_clock_get_cycles+0x9/0x10 [ 80.145172] [<ffffffff810afd1e>] ? __getnstimeofday+0x3e/0xd0 [ 80.145172] [<ffffffff8152509e>] SyS_sendto+0xe/0x10 [ 80.145172] [<ffffffff8164efd9>] system_call_fastpath+0x16/0x1b [ 80.145172] Code: fe ff ff 41 f6 45 2a 06 0f 85 ca fe ff ff 49 8b 7e 08 4c 89 ee e8 94 ef ff ff e9 b9 fe ff ff 48 8b 82 28 05 00 00 e9 01 ff ff ff <0f> 0b 49 8b 54 24 30 0d 00 00 40 00 89 83 14 01 00 00 48 89 53 [ 80.145172] RIP [<ffffffff815f3b5d>] fib6_add+0x75d/0x830 [ 80.145172] RSP <ffff880118771798> [ 80.387413] ---[ end trace 02f20b7a8b81ed95 ]--- [ 80.390154] Kernel panic - not syncing: Fatal exception in interrupt Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com> Cc: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-10 21:00:57 +00:00
}
}
}
static void rt6_do_update_pmtu(struct rt6_info *rt, u32 mtu)
{
struct net *net = dev_net(rt->dst.dev);
rt->rt6i_flags |= RTF_MODIFIED;
rt->rt6i_pmtu = mtu;
rt6_update_expires(rt, net->ipv6.sysctl.ip6_rt_mtu_expires);
}
static void __ip6_rt_update_pmtu(struct dst_entry *dst, const struct sock *sk,
const struct ipv6hdr *iph, u32 mtu)
{
struct rt6_info *rt6 = (struct rt6_info *)dst;
if (rt6->rt6i_flags & RTF_LOCAL)
return;
dst_confirm(dst);
mtu = max_t(u32, mtu, IPV6_MIN_MTU);
if (mtu >= dst_mtu(dst))
return;
if (rt6->rt6i_flags & RTF_CACHE) {
rt6_do_update_pmtu(rt6, mtu);
} else {
const struct in6_addr *daddr, *saddr;
struct rt6_info *nrt6;
if (iph) {
daddr = &iph->daddr;
saddr = &iph->saddr;
} else if (sk) {
daddr = &sk->sk_v6_daddr;
saddr = &inet6_sk(sk)->saddr;
} else {
return;
}
nrt6 = ip6_rt_cache_alloc(rt6, daddr, saddr);
if (nrt6) {
rt6_do_update_pmtu(nrt6, mtu);
/* ip6_ins_rt(nrt6) will bump the
* rt6->rt6i_node->fn_sernum
* which will fail the next rt6_check() and
* invalidate the sk->sk_dst_cache.
*/
ip6_ins_rt(nrt6);
}
}
}
static void ip6_rt_update_pmtu(struct dst_entry *dst, struct sock *sk,
struct sk_buff *skb, u32 mtu)
{
__ip6_rt_update_pmtu(dst, sk, skb ? ipv6_hdr(skb) : NULL, mtu);
}
void ip6_update_pmtu(struct sk_buff *skb, struct net *net, __be32 mtu,
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 ? mark : IP6_REPLY_MARK(net, skb->mark);
fl6.daddr = iph->daddr;
fl6.saddr = iph->saddr;
fl6.flowlabel = ip6_flowinfo(iph);
dst = ip6_route_output(net, NULL, &fl6);
if (!dst->error)
__ip6_rt_update_pmtu(dst, NULL, iph, ntohl(mtu));
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);
/* 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;
}
if (rt == net->ipv6.ip6_null_entry) {
fn = fib6_backtrack(fn, &fl6->saddr);
if (fn)
goto restart;
}
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);
}
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_iif = LOOPBACK_IFINDEX;
fl6.flowi6_oif = oif;
fl6.flowi6_mark = mark;
fl6.daddr = iph->daddr;
fl6.saddr = iph->saddr;
fl6.flowlabel = ip6_flowinfo(iph);
dst = ip6_route_redirect(net, &fl6, &ipv6_hdr(skb)->saddr);
rt6_do_redirect(dst, NULL, skb);
dst_release(dst);
}
EXPORT_SYMBOL_GPL(ip6_redirect);
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_iif = LOOPBACK_IFINDEX;
fl6.flowi6_oif = oif;
fl6.flowi6_mark = mark;
fl6.daddr = msg->dest;
fl6.saddr = iph->daddr;
dst = ip6_route_redirect(net, &fl6, &iph->saddr);
rt6_do_redirect(dst, NULL, skb);
dst_release(dst);
}
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);
static unsigned int ip6_default_advmss(const struct dst_entry *dst)
{
struct net_device *dev = dst->dev;
unsigned int mtu = dst_mtu(dst);
struct net *net = dev_net(dev);
mtu -= sizeof(struct ipv6hdr) + sizeof(struct tcphdr);
if (mtu < net->ipv6.sysctl.ip6_rt_min_advmss)
mtu = net->ipv6.sysctl.ip6_rt_min_advmss;
/*
* 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,
* rely only on pmtu discovery"
*/
if (mtu > IPV6_MAXPLEN - sizeof(struct tcphdr))
mtu = IPV6_MAXPLEN;
return mtu;
}
static unsigned int ip6_mtu(const struct dst_entry *dst)
{
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
const struct rt6_info *rt = (const struct rt6_info *)dst;
unsigned int mtu = rt->rt6i_pmtu;
struct inet6_dev *idev;
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
if (mtu)
goto out;
mtu = dst_metric_raw(dst, RTAX_MTU);
if (mtu)
goto out;
mtu = IPV6_MIN_MTU;
rcu_read_lock();
idev = __in6_dev_get(dst->dev);
if (idev)
mtu = idev->cnf.mtu6;
rcu_read_unlock();
out:
return min_t(unsigned int, mtu, IP6_MAX_MTU);
}
static struct dst_entry *icmp6_dst_gc_list;
static DEFINE_SPINLOCK(icmp6_dst_lock);
struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
struct flowi6 *fl6)
{
struct dst_entry *dst;
struct rt6_info *rt;
struct inet6_dev *idev = in6_dev_get(dev);
struct net *net = dev_net(dev);
if (unlikely(!idev))
return ERR_PTR(-ENODEV);
rt = ip6_dst_alloc(net, dev, 0);
if (unlikely(!rt)) {
in6_dev_put(idev);
dst = ERR_PTR(-ENOMEM);
goto out;
}
rt->dst.flags |= DST_HOST;
rt->dst.output = ip6_output;
atomic_set(&rt->dst.__refcnt, 1);
rt->rt6i_gateway = fl6->daddr;
rt->rt6i_dst.addr = fl6->daddr;
rt->rt6i_dst.plen = 128;
rt->rt6i_idev = idev;
dst_metric_set(&rt->dst, RTAX_HOPLIMIT, 0);
spin_lock_bh(&icmp6_dst_lock);
rt->dst.next = icmp6_dst_gc_list;
icmp6_dst_gc_list = &rt->dst;
spin_unlock_bh(&icmp6_dst_lock);
fib6_force_start_gc(net);
dst = xfrm_lookup(net, &rt->dst, flowi6_to_flowi(fl6), NULL, 0);
out:
return dst;
}
int icmp6_dst_gc(void)
{
struct dst_entry *dst, **pprev;
int more = 0;
spin_lock_bh(&icmp6_dst_lock);
pprev = &icmp6_dst_gc_list;
while ((dst = *pprev) != NULL) {
if (!atomic_read(&dst->__refcnt)) {
*pprev = dst->next;
dst_free(dst);
} else {
pprev = &dst->next;
++more;
}
}
spin_unlock_bh(&icmp6_dst_lock);
return more;
}
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);
}
static int ip6_dst_gc(struct dst_ops *ops)
{
struct net *net = container_of(ops, struct net, ipv6.ip6_dst_ops);
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;
int entries;
entries = dst_entries_get_fast(ops);
if (time_after(rt_last_gc + rt_min_interval, jiffies) &&
entries <= rt_max_size)
goto out;
net->ipv6.ip6_rt_gc_expire++;
fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net, true);
entries = dst_entries_get_slow(ops);
if (entries < ops->gc_thresh)
net->ipv6.ip6_rt_gc_expire = rt_gc_timeout>>1;
out:
net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>rt_elasticity;
return entries > rt_max_size;
}
static int ip6_convert_metrics(struct mx6_config *mxc,
const struct fib6_config *cfg)
{
bool ecn_ca = false;
struct nlattr *nla;
int remaining;
u32 *mp;
if (!cfg->fc_mx)
return 0;
mp = kzalloc(sizeof(u32) * RTAX_MAX, GFP_KERNEL);
if (unlikely(!mp))
return -ENOMEM;
nla_for_each_attr(nla, cfg->fc_mx, cfg->fc_mx_len, remaining) {
int type = nla_type(nla);
u32 val;
if (!type)
continue;
if (unlikely(type > RTAX_MAX))
goto err;
if (type == RTAX_CC_ALGO) {
char tmp[TCP_CA_NAME_MAX];
nla_strlcpy(tmp, nla, sizeof(tmp));
val = tcp_ca_get_key_by_name(tmp, &ecn_ca);
if (val == TCP_CA_UNSPEC)
goto err;
} else {
val = nla_get_u32(nla);
}
if (type == RTAX_FEATURES && (val & ~RTAX_FEATURE_MASK))
goto err;
mp[type - 1] = val;
__set_bit(type - 1, mxc->mx_valid);
}
if (ecn_ca) {
__set_bit(RTAX_FEATURES - 1, mxc->mx_valid);
mp[RTAX_FEATURES - 1] |= DST_FEATURE_ECN_CA;
}
mxc->mx = mp;
return 0;
err:
kfree(mp);
return -EINVAL;
}
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
int ip6_route_info_create(struct fib6_config *cfg, struct rt6_info **rt_ret)
{
int err;
struct net *net = cfg->fc_nlinfo.nl_net;
struct rt6_info *rt = NULL;
struct net_device *dev = NULL;
struct inet6_dev *idev = NULL;
struct fib6_table *table;
int addr_type;
if (cfg->fc_dst_len > 128 || cfg->fc_src_len > 128)
return -EINVAL;
#ifndef CONFIG_IPV6_SUBTREES
if (cfg->fc_src_len)
return -EINVAL;
#endif
if (cfg->fc_ifindex) {
err = -ENODEV;
dev = dev_get_by_index(net, cfg->fc_ifindex);
if (!dev)
goto out;
idev = in6_dev_get(dev);
if (!idev)
goto out;
}
if (cfg->fc_metric == 0)
cfg->fc_metric = IP6_RT_PRIO_USER;
err = -ENOBUFS;
if (cfg->fc_nlinfo.nlh &&
!(cfg->fc_nlinfo.nlh->nlmsg_flags & NLM_F_CREATE)) {
table = fib6_get_table(net, cfg->fc_table);
if (!table) {
pr_warn("NLM_F_CREATE should be specified when creating new route\n");
table = fib6_new_table(net, cfg->fc_table);
}
} else {
table = fib6_new_table(net, cfg->fc_table);
}
if (!table)
goto out;
rt = ip6_dst_alloc(net, NULL,
(cfg->fc_flags & RTF_ADDRCONF) ? 0 : DST_NOCOUNT);
if (!rt) {
err = -ENOMEM;
goto out;
}
if (cfg->fc_flags & RTF_EXPIRES)
rt6_set_expires(rt, jiffies +
clock_t_to_jiffies(cfg->fc_expires));
else
rt6_clean_expires(rt);
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);
if (addr_type & IPV6_ADDR_MULTICAST)
rt->dst.input = ip6_mc_input;
else if (cfg->fc_flags & RTF_LOCAL)
rt->dst.input = ip6_input;
else
rt->dst.input = ip6_forward;
rt->dst.output = ip6_output;
if (cfg->fc_encap) {
struct lwtunnel_state *lwtstate;
err = lwtunnel_build_state(dev, cfg->fc_encap_type,
cfg->fc_encap, AF_INET6, cfg,
&lwtstate);
if (err)
goto out;
rt->dst.lwtstate = lwtstate_get(lwtstate);
if (lwtunnel_output_redirect(rt->dst.lwtstate)) {
rt->dst.lwtstate->orig_output = rt->dst.output;
rt->dst.output = lwtunnel_output;
}
if (lwtunnel_input_redirect(rt->dst.lwtstate)) {
rt->dst.lwtstate->orig_input = rt->dst.input;
rt->dst.input = lwtunnel_input;
}
}
ipv6_addr_prefix(&rt->rt6i_dst.addr, &cfg->fc_dst, cfg->fc_dst_len);
rt->rt6i_dst.plen = cfg->fc_dst_len;
if (rt->rt6i_dst.plen == 128)
ipv6: do not overwrite inetpeer metrics prematurely If an IPv6 host route with metrics exists, an attempt to add a new route for the same target with different metrics fails but rewrites the metrics anyway: 12sp0:~ # ip route add fec0::1 dev eth0 rto_min 1000 12sp0:~ # ip -6 route show fe80::/64 dev eth0 proto kernel metric 256 fec0::1 dev eth0 metric 1024 rto_min lock 1s 12sp0:~ # ip route add fec0::1 dev eth0 rto_min 1500 RTNETLINK answers: File exists 12sp0:~ # ip -6 route show fe80::/64 dev eth0 proto kernel metric 256 fec0::1 dev eth0 metric 1024 rto_min lock 1.5s This is caused by all IPv6 host routes using the metrics in their inetpeer (or the shared default). This also holds for the new route created in ip6_route_add() which shares the metrics with the already existing route and thus ip6_route_add() rewrites the metrics even if the new route ends up not being used at all. Another problem is that old metrics in inetpeer can reappear unexpectedly for a new route, e.g. 12sp0:~ # ip route add fec0::1 dev eth0 rto_min 1000 12sp0:~ # ip route del fec0::1 12sp0:~ # ip route add fec0::1 dev eth0 12sp0:~ # ip route change fec0::1 dev eth0 hoplimit 10 12sp0:~ # ip -6 route show fe80::/64 dev eth0 proto kernel metric 256 fec0::1 dev eth0 metric 1024 hoplimit 10 rto_min lock 1s Resolve the first problem by moving the setting of metrics down into fib6_add_rt2node() to the point we are sure we are inserting the new route into the tree. Second problem is addressed by introducing new flag DST_METRICS_FORCE_OVERWRITE which is set for a new host route in ip6_route_add() and makes ipv6_cow_metrics() always overwrite the metrics in inetpeer (even if they are not "new"); it is reset after that. v5: use a flag in _metrics member rather than one in flags v4: fix a typo making a condition always true (thanks to Hannes Frederic Sowa) v3: rewritten based on David Miller's idea to move setting the metrics (and allocation in non-host case) down to the point we already know the route is to be inserted. Also rebased to net-next as it is quite late in the cycle. Signed-off-by: Michal Kubecek <mkubecek@suse.cz> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-27 12:04:08 +00:00
rt->dst.flags |= DST_HOST;
#ifdef CONFIG_IPV6_SUBTREES
ipv6_addr_prefix(&rt->rt6i_src.addr, &cfg->fc_src, cfg->fc_src_len);
rt->rt6i_src.plen = cfg->fc_src_len;
#endif
rt->rt6i_metric = cfg->fc_metric;
/* We cannot add true routes via loopback here,
they would result in kernel looping; promote them to reject routes
*/
if ((cfg->fc_flags & RTF_REJECT) ||
(dev && (dev->flags & IFF_LOOPBACK) &&
!(addr_type & IPV6_ADDR_LOOPBACK) &&
!(cfg->fc_flags & RTF_LOCAL))) {
/* hold loopback dev/idev if we haven't done so. */
if (dev != net->loopback_dev) {
if (dev) {
dev_put(dev);
in6_dev_put(idev);
}
dev = net->loopback_dev;
dev_hold(dev);
idev = in6_dev_get(dev);
if (!idev) {
err = -ENODEV;
goto out;
}
}
rt->rt6i_flags = RTF_REJECT|RTF_NONEXTHOP;
switch (cfg->fc_type) {
case RTN_BLACKHOLE:
rt->dst.error = -EINVAL;
rt->dst.output = dst_discard_sk;
rt->dst.input = dst_discard;
break;
case RTN_PROHIBIT:
rt->dst.error = -EACCES;
rt->dst.output = ip6_pkt_prohibit_out;
rt->dst.input = ip6_pkt_prohibit;
break;
case RTN_THROW:
case RTN_UNREACHABLE:
default:
rt->dst.error = (cfg->fc_type == RTN_THROW) ? -EAGAIN
: (cfg->fc_type == RTN_UNREACHABLE)
? -EHOSTUNREACH : -ENETUNREACH;
rt->dst.output = ip6_pkt_discard_out;
rt->dst.input = ip6_pkt_discard;
break;
}
goto install_route;
}
if (cfg->fc_flags & RTF_GATEWAY) {
const struct in6_addr *gw_addr;
int gwa_type;
gw_addr = &cfg->fc_gateway;
gwa_type = ipv6_addr_type(gw_addr);
/* if gw_addr is local we will fail to detect this in case
* address is still TENTATIVE (DAD in progress). rt6_lookup()
* will return already-added prefix route via interface that
* prefix route was assigned to, which might be non-loopback.
*/
err = -EINVAL;
if (ipv6_chk_addr_and_flags(net, gw_addr,
gwa_type & IPV6_ADDR_LINKLOCAL ?
dev : NULL, 0, 0))
goto out;
rt->rt6i_gateway = *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
*/
if (!(gwa_type & IPV6_ADDR_UNICAST))
goto out;
grt = rt6_lookup(net, gw_addr, NULL, cfg->fc_ifindex, 1);
err = -EHOSTUNREACH;
if (!grt)
goto out;
if (dev) {
if (dev != grt->dst.dev) {
ip6_rt_put(grt);
goto out;
}
} else {
dev = grt->dst.dev;
idev = grt->rt6i_idev;
dev_hold(dev);
in6_dev_hold(grt->rt6i_idev);
}
if (!(grt->rt6i_flags & RTF_GATEWAY))
err = 0;
ip6_rt_put(grt);
if (err)
goto out;
}
err = -EINVAL;
if (!dev || (dev->flags & IFF_LOOPBACK))
goto out;
}
err = -ENODEV;
if (!dev)
goto out;
if (!ipv6_addr_any(&cfg->fc_prefsrc)) {
if (!ipv6_chk_addr(net, &cfg->fc_prefsrc, dev, 0)) {
err = -EINVAL;
goto out;
}
rt->rt6i_prefsrc.addr = cfg->fc_prefsrc;
rt->rt6i_prefsrc.plen = 128;
} else
rt->rt6i_prefsrc.plen = 0;
rt->rt6i_flags = cfg->fc_flags;
install_route:
rt->dst.dev = dev;
rt->rt6i_idev = idev;
rt->rt6i_table = table;
cfg->fc_nlinfo.nl_net = dev_net(dev);
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
*rt_ret = rt;
return 0;
out:
if (dev)
dev_put(dev);
if (idev)
in6_dev_put(idev);
if (rt)
dst_free(&rt->dst);
*rt_ret = NULL;
return err;
}
int ip6_route_add(struct fib6_config *cfg)
{
struct mx6_config mxc = { .mx = NULL, };
struct rt6_info *rt = NULL;
int err;
err = ip6_route_info_create(cfg, &rt);
if (err)
goto out;
err = ip6_convert_metrics(&mxc, cfg);
if (err)
goto out;
err = __ip6_ins_rt(rt, &cfg->fc_nlinfo, &mxc);
kfree(mxc.mx);
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
return err;
out:
if (rt)
dst_free(&rt->dst);
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
return err;
}
static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
{
int err;
struct fib6_table *table;
struct net *net = dev_net(rt->dst.dev);
if (rt == net->ipv6.ip6_null_entry ||
rt->dst.flags & DST_NOCACHE) {
err = -ENOENT;
goto out;
}
table = rt->rt6i_table;
write_lock_bh(&table->tb6_lock);
err = fib6_del(rt, info);
write_unlock_bh(&table->tb6_lock);
out:
ip6_rt_put(rt);
return err;
}
int ip6_del_rt(struct rt6_info *rt)
{
struct nl_info info = {
.nl_net = dev_net(rt->dst.dev),
};
return __ip6_del_rt(rt, &info);
}
static int ip6_route_del(struct fib6_config *cfg)
{
struct fib6_table *table;
struct fib6_node *fn;
struct rt6_info *rt;
int err = -ESRCH;
table = fib6_get_table(cfg->fc_nlinfo.nl_net, cfg->fc_table);
if (!table)
return err;
read_lock_bh(&table->tb6_lock);
fn = fib6_locate(&table->tb6_root,
&cfg->fc_dst, cfg->fc_dst_len,
&cfg->fc_src, cfg->fc_src_len);
if (fn) {
for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
if ((rt->rt6i_flags & RTF_CACHE) &&
!(cfg->fc_flags & RTF_CACHE))
continue;
if (cfg->fc_ifindex &&
(!rt->dst.dev ||
rt->dst.dev->ifindex != cfg->fc_ifindex))
continue;
if (cfg->fc_flags & RTF_GATEWAY &&
!ipv6_addr_equal(&cfg->fc_gateway, &rt->rt6i_gateway))
continue;
if (cfg->fc_metric && cfg->fc_metric != rt->rt6i_metric)
continue;
dst_hold(&rt->dst);
read_unlock_bh(&table->tb6_lock);
return __ip6_del_rt(rt, &cfg->fc_nlinfo);
}
}
read_unlock_bh(&table->tb6_lock);
return err;
}
static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_buff *skb)
{
struct net *net = dev_net(skb->dev);
struct netevent_redirect netevent;
struct rt6_info *rt, *nrt = NULL;
struct ndisc_options ndopts;
struct inet6_dev *in6_dev;
struct neighbour *neigh;
struct rd_msg *msg;
int optlen, on_link;
u8 *lladdr;
optlen = skb_tail_pointer(skb) - skb_transport_header(skb);
optlen -= sizeof(*msg);
if (optlen < 0) {
net_dbg_ratelimited("rt6_do_redirect: packet too short\n");
return;
}
msg = (struct rd_msg *)icmp6_hdr(skb);
if (ipv6_addr_is_multicast(&msg->dest)) {
net_dbg_ratelimited("rt6_do_redirect: destination address is multicast\n");
return;
}
on_link = 0;
if (ipv6_addr_equal(&msg->dest, &msg->target)) {
on_link = 1;
} else if (ipv6_addr_type(&msg->target) !=
(IPV6_ADDR_UNICAST|IPV6_ADDR_LINKLOCAL)) {
net_dbg_ratelimited("rt6_do_redirect: target address is not link-local unicast\n");
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.
*/
if (!ndisc_parse_options(msg->opt, optlen, &ndopts)) {
net_dbg_ratelimited("rt6_redirect: invalid ND options\n");
return;
}
lladdr = NULL;
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;
}
}
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");
return;
}
/* 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);
neigh = __neigh_lookup(&nd_tbl, &msg->target, skb->dev, 1);
if (!neigh)
return;
/*
* We have finally decided to accept it.
*/
neigh_update(neigh, lladdr, NUD_STALE,
NEIGH_UPDATE_F_WEAK_OVERRIDE|
NEIGH_UPDATE_F_OVERRIDE|
(on_link ? 0 : (NEIGH_UPDATE_F_OVERRIDE_ISROUTER|
NEIGH_UPDATE_F_ISROUTER))
);
nrt = ip6_rt_cache_alloc(rt, &msg->dest, NULL);
if (!nrt)
goto out;
nrt->rt6i_flags = RTF_GATEWAY|RTF_UP|RTF_DYNAMIC|RTF_CACHE;
if (on_link)
nrt->rt6i_flags &= ~RTF_GATEWAY;
nrt->rt6i_gateway = *(struct in6_addr *)neigh->primary_key;
if (ip6_ins_rt(nrt))
goto out;
netevent.old = &rt->dst;
netevent.new = &nrt->dst;
netevent.daddr = &msg->dest;
netevent.neigh = neigh;
call_netevent_notifiers(NETEVENT_REDIRECT, &netevent);
if (rt->rt6i_flags & RTF_CACHE) {
rt = (struct rt6_info *) dst_clone(&rt->dst);
ip6_del_rt(rt);
}
out:
neigh_release(neigh);
}
/*
* Misc support functions
*/
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
static void rt6_set_from(struct rt6_info *rt, struct rt6_info *from)
{
BUG_ON(from->dst.from);
rt->rt6i_flags &= ~RTF_EXPIRES;
dst_hold(&from->dst);
rt->dst.from = &from->dst;
dst_init_metrics(&rt->dst, dst_metrics_ptr(&from->dst), true);
}
static void ip6_rt_copy_init(struct rt6_info *rt, struct rt6_info *ort)
{
rt->dst.input = ort->dst.input;
rt->dst.output = ort->dst.output;
rt->rt6i_dst = ort->rt6i_dst;
rt->dst.error = ort->dst.error;
rt->rt6i_idev = ort->rt6i_idev;
if (rt->rt6i_idev)
in6_dev_hold(rt->rt6i_idev);
rt->dst.lastuse = jiffies;
rt->rt6i_gateway = ort->rt6i_gateway;
rt->rt6i_flags = ort->rt6i_flags;
rt6_set_from(rt, ort);
rt->rt6i_metric = ort->rt6i_metric;
#ifdef CONFIG_IPV6_SUBTREES
rt->rt6i_src = ort->rt6i_src;
#endif
rt->rt6i_prefsrc = ort->rt6i_prefsrc;
rt->rt6i_table = ort->rt6i_table;
rt->dst.lwtstate = lwtstate_get(ort->dst.lwtstate);
}
#ifdef CONFIG_IPV6_ROUTE_INFO
static struct rt6_info *rt6_get_route_info(struct net *net,
const struct in6_addr *prefix, int prefixlen,
const struct in6_addr *gwaddr, int ifindex)
{
struct fib6_node *fn;
struct rt6_info *rt = NULL;
struct fib6_table *table;
table = fib6_get_table(net, RT6_TABLE_INFO);
if (!table)
return NULL;
read_lock_bh(&table->tb6_lock);
fn = fib6_locate(&table->tb6_root, prefix, prefixlen, NULL, 0);
if (!fn)
goto out;
for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
if (rt->dst.dev->ifindex != ifindex)
continue;
if ((rt->rt6i_flags & (RTF_ROUTEINFO|RTF_GATEWAY)) != (RTF_ROUTEINFO|RTF_GATEWAY))
continue;
if (!ipv6_addr_equal(&rt->rt6i_gateway, gwaddr))
continue;
dst_hold(&rt->dst);
break;
}
out:
read_unlock_bh(&table->tb6_lock);
return rt;
}
static struct rt6_info *rt6_add_route_info(struct net *net,
const struct in6_addr *prefix, int prefixlen,
const struct in6_addr *gwaddr, int ifindex,
unsigned int pref)
{
struct fib6_config cfg = {
.fc_table = RT6_TABLE_INFO,
.fc_metric = IP6_RT_PRIO_USER,
.fc_ifindex = ifindex,
.fc_dst_len = prefixlen,
.fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO |
RTF_UP | RTF_PREF(pref),
.fc_nlinfo.portid = 0,
.fc_nlinfo.nlh = NULL,
.fc_nlinfo.nl_net = net,
};
cfg.fc_dst = *prefix;
cfg.fc_gateway = *gwaddr;
/* We should treat it as a default route if prefix length is 0. */
if (!prefixlen)
cfg.fc_flags |= RTF_DEFAULT;
ip6_route_add(&cfg);
return rt6_get_route_info(net, prefix, prefixlen, gwaddr, ifindex);
}
#endif
struct rt6_info *rt6_get_dflt_router(const struct in6_addr *addr, struct net_device *dev)
{
struct rt6_info *rt;
struct fib6_table *table;
table = fib6_get_table(dev_net(dev), RT6_TABLE_DFLT);
if (!table)
return NULL;
read_lock_bh(&table->tb6_lock);
for (rt = table->tb6_root.leaf; rt; rt = rt->dst.rt6_next) {
if (dev == rt->dst.dev &&
((rt->rt6i_flags & (RTF_ADDRCONF | RTF_DEFAULT)) == (RTF_ADDRCONF | RTF_DEFAULT)) &&
ipv6_addr_equal(&rt->rt6i_gateway, addr))
break;
}
if (rt)
dst_hold(&rt->dst);
read_unlock_bh(&table->tb6_lock);
return rt;
}
struct rt6_info *rt6_add_dflt_router(const struct in6_addr *gwaddr,
struct net_device *dev,
unsigned int pref)
{
struct fib6_config cfg = {
.fc_table = RT6_TABLE_DFLT,
.fc_metric = IP6_RT_PRIO_USER,
.fc_ifindex = dev->ifindex,
.fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT |
RTF_UP | RTF_EXPIRES | RTF_PREF(pref),
.fc_nlinfo.portid = 0,
.fc_nlinfo.nlh = NULL,
.fc_nlinfo.nl_net = dev_net(dev),
};
cfg.fc_gateway = *gwaddr;
ip6_route_add(&cfg);
return rt6_get_dflt_router(gwaddr, dev);
}
void rt6_purge_dflt_routers(struct net *net)
{
struct rt6_info *rt;
struct fib6_table *table;
/* NOTE: Keep consistent with rt6_get_dflt_router */
table = fib6_get_table(net, RT6_TABLE_DFLT);
if (!table)
return;
restart:
read_lock_bh(&table->tb6_lock);
for (rt = table->tb6_root.leaf; rt; rt = rt->dst.rt6_next) {
if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF) &&
(!rt->rt6i_idev || rt->rt6i_idev->cnf.accept_ra != 2)) {
dst_hold(&rt->dst);
read_unlock_bh(&table->tb6_lock);
ip6_del_rt(rt);
goto restart;
}
}
read_unlock_bh(&table->tb6_lock);
}
static void rtmsg_to_fib6_config(struct net *net,
struct in6_rtmsg *rtmsg,
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;
cfg->fc_nlinfo.nl_net = net;
cfg->fc_dst = rtmsg->rtmsg_dst;
cfg->fc_src = rtmsg->rtmsg_src;
cfg->fc_gateway = rtmsg->rtmsg_gateway;
}
int ipv6_route_ioctl(struct net *net, unsigned int cmd, void __user *arg)
{
struct fib6_config cfg;
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>
2012-11-16 03:03:06 +00:00
if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
return -EPERM;
err = copy_from_user(&rtmsg, arg,
sizeof(struct in6_rtmsg));
if (err)
return -EFAULT;
rtmsg_to_fib6_config(net, &rtmsg, &cfg);
rtnl_lock();
switch (cmd) {
case SIOCADDRT:
err = ip6_route_add(&cfg);
break;
case SIOCDELRT:
err = ip6_route_del(&cfg);
break;
default:
err = -EINVAL;
}
rtnl_unlock();
return err;
}
return -EINVAL;
}
/*
* Drop the packet on the floor
*/
static int ip6_pkt_drop(struct sk_buff *skb, u8 code, int ipstats_mib_noroutes)
{
int type;
struct dst_entry *dst = skb_dst(skb);
switch (ipstats_mib_noroutes) {
case IPSTATS_MIB_INNOROUTES:
type = ipv6_addr_type(&ipv6_hdr(skb)->daddr);
if (type == IPV6_ADDR_ANY) {
IP6_INC_STATS(dev_net(dst->dev), ip6_dst_idev(dst),
IPSTATS_MIB_INADDRERRORS);
break;
}
/* FALLTHROUGH */
case IPSTATS_MIB_OUTNOROUTES:
IP6_INC_STATS(dev_net(dst->dev), ip6_dst_idev(dst),
ipstats_mib_noroutes);
break;
}
icmpv6_send(skb, ICMPV6_DEST_UNREACH, code, 0);
kfree_skb(skb);
return 0;
}
static int ip6_pkt_discard(struct sk_buff *skb)
{
return ip6_pkt_drop(skb, ICMPV6_NOROUTE, IPSTATS_MIB_INNOROUTES);
}
static int ip6_pkt_discard_out(struct sock *sk, struct sk_buff *skb)
{
skb->dev = skb_dst(skb)->dev;
return ip6_pkt_drop(skb, ICMPV6_NOROUTE, IPSTATS_MIB_OUTNOROUTES);
}
static int ip6_pkt_prohibit(struct sk_buff *skb)
{
return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED, IPSTATS_MIB_INNOROUTES);
}
static int ip6_pkt_prohibit_out(struct sock *sk, struct sk_buff *skb)
{
skb->dev = skb_dst(skb)->dev;
return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED, IPSTATS_MIB_OUTNOROUTES);
}
/*
* Allocate a dst for local (unicast / anycast) address.
*/
struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
const struct in6_addr *addr,
bool anycast)
{
struct net *net = dev_net(idev->dev);
struct rt6_info *rt = ip6_dst_alloc(net, net->loopback_dev,
DST_NOCOUNT);
if (!rt)
return ERR_PTR(-ENOMEM);
in6_dev_hold(idev);
rt->dst.flags |= DST_HOST;
rt->dst.input = ip6_input;
rt->dst.output = ip6_output;
rt->rt6i_idev = idev;
rt->rt6i_flags = RTF_UP | RTF_NONEXTHOP;
if (anycast)
rt->rt6i_flags |= RTF_ANYCAST;
else
rt->rt6i_flags |= RTF_LOCAL;
rt->rt6i_gateway = *addr;
rt->rt6i_dst.addr = *addr;
rt->rt6i_dst.plen = 128;
rt->rt6i_table = fib6_get_table(net, RT6_TABLE_LOCAL);
rt->dst.flags |= DST_NOCACHE;
atomic_set(&rt->dst.__refcnt, 1);
return rt;
}
int ip6_route_get_saddr(struct net *net,
struct rt6_info *rt,
const struct in6_addr *daddr,
unsigned int prefs,
struct in6_addr *saddr)
{
struct inet6_dev *idev =
rt ? ip6_dst_idev((struct dst_entry *)rt) : NULL;
int err = 0;
if (rt && rt->rt6i_prefsrc.plen)
*saddr = rt->rt6i_prefsrc.addr;
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;
if (((void *)rt->dst.dev == dev || !dev) &&
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, &adni);
}
#define RTF_RA_ROUTER (RTF_ADDRCONF | RTF_DEFAULT | RTF_GATEWAY)
#define RTF_CACHE_GATEWAY (RTF_GATEWAY | RTF_CACHE)
/* Remove routers and update dst entries when gateway turn into host. */
static int fib6_clean_tohost(struct rt6_info *rt, void *arg)
{
struct in6_addr *gateway = (struct in6_addr *)arg;
if ((((rt->rt6i_flags & RTF_RA_ROUTER) == RTF_RA_ROUTER) ||
((rt->rt6i_flags & RTF_CACHE_GATEWAY) == RTF_CACHE_GATEWAY)) &&
ipv6_addr_equal(gateway, &rt->rt6i_gateway)) {
return -1;
}
return 0;
}
void rt6_clean_tohost(struct net *net, struct in6_addr *gateway)
{
fib6_clean_all(net, fib6_clean_tohost, gateway);
}
struct arg_dev_net {
struct net_device *dev;
struct net *net;
};
static int fib6_ifdown(struct rt6_info *rt, void *arg)
{
const struct arg_dev_net *adn = arg;
const struct net_device *dev = adn->dev;
if ((rt->dst.dev == dev || !dev) &&
rt != adn->net->ipv6.ip6_null_entry)
return -1;
return 0;
}
void rt6_ifdown(struct net *net, struct net_device *dev)
{
struct arg_dev_net adn = {
.dev = dev,
.net = net,
};
fib6_clean_all(net, fib6_ifdown, &adn);
icmp6_clean_all(fib6_ifdown, &adn);
if (dev)
rt6_uncached_list_flush_dev(net, dev);
}
struct rt6_mtu_change_arg {
struct net_device *dev;
unsigned int mtu;
};
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);
if (!idev)
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.
*/
if (rt->dst.dev == arg->dev &&
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
!dst_metric_locked(&rt->dst, RTAX_MTU)) {
if (rt->rt6i_flags & RTF_CACHE) {
/* For RTF_CACHE with rt6i_pmtu == 0
* (i.e. a redirected route),
* the metrics of its rt->dst.from has already
* been updated.
*/
if (rt->rt6i_pmtu && rt->rt6i_pmtu > arg->mtu)
rt->rt6i_pmtu = arg->mtu;
} else if (dst_mtu(&rt->dst) >= arg->mtu ||
(dst_mtu(&rt->dst) < arg->mtu &&
dst_mtu(&rt->dst) == idev->cnf.mtu6)) {
dst_metric_set(&rt->dst, RTAX_MTU, arg->mtu);
}
}
return 0;
}
void rt6_mtu_change(struct net_device *dev, unsigned int mtu)
{
struct rt6_mtu_change_arg arg = {
.dev = dev,
.mtu = mtu,
};
fib6_clean_all(dev_net(dev), rt6_mtu_change_route, &arg);
}
static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = {
[RTA_GATEWAY] = { .len = sizeof(struct in6_addr) },
[RTA_OIF] = { .type = NLA_U32 },
[RTA_IIF] = { .type = NLA_U32 },
[RTA_PRIORITY] = { .type = NLA_U32 },
[RTA_METRICS] = { .type = NLA_NESTED },
[RTA_MULTIPATH] = { .len = sizeof(struct rtnexthop) },
[RTA_PREF] = { .type = NLA_U8 },
[RTA_ENCAP_TYPE] = { .type = NLA_U16 },
[RTA_ENCAP] = { .type = NLA_NESTED },
};
static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh,
struct fib6_config *cfg)
{
struct rtmsg *rtm;
struct nlattr *tb[RTA_MAX+1];
unsigned int pref;
int err;
err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv6_policy);
if (err < 0)
goto errout;
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;
cfg->fc_type = rtm->rtm_type;
if (rtm->rtm_type == RTN_UNREACHABLE ||
rtm->rtm_type == RTN_BLACKHOLE ||
rtm->rtm_type == RTN_PROHIBIT ||
rtm->rtm_type == RTN_THROW)
cfg->fc_flags |= RTF_REJECT;
if (rtm->rtm_type == RTN_LOCAL)
cfg->fc_flags |= RTF_LOCAL;
if (rtm->rtm_flags & RTM_F_CLONED)
cfg->fc_flags |= RTF_CACHE;
cfg->fc_nlinfo.portid = NETLINK_CB(skb).portid;
cfg->fc_nlinfo.nlh = nlh;
cfg->fc_nlinfo.nl_net = sock_net(skb->sk);
if (tb[RTA_GATEWAY]) {
cfg->fc_gateway = nla_get_in6_addr(tb[RTA_GATEWAY]);
cfg->fc_flags |= RTF_GATEWAY;
}
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);
}
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);
}
if (tb[RTA_PREFSRC])
cfg->fc_prefsrc = nla_get_in6_addr(tb[RTA_PREFSRC]);
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]);
}
if (tb[RTA_TABLE])
cfg->fc_table = nla_get_u32(tb[RTA_TABLE]);
if (tb[RTA_MULTIPATH]) {
cfg->fc_mp = nla_data(tb[RTA_MULTIPATH]);
cfg->fc_mp_len = nla_len(tb[RTA_MULTIPATH]);
}
if (tb[RTA_PREF]) {
pref = nla_get_u8(tb[RTA_PREF]);
if (pref != ICMPV6_ROUTER_PREF_LOW &&
pref != ICMPV6_ROUTER_PREF_HIGH)
pref = ICMPV6_ROUTER_PREF_MEDIUM;
cfg->fc_flags |= RTF_PREF(pref);
}
if (tb[RTA_ENCAP])
cfg->fc_encap = tb[RTA_ENCAP];
if (tb[RTA_ENCAP_TYPE])
cfg->fc_encap_type = nla_get_u16(tb[RTA_ENCAP_TYPE]);
err = 0;
errout:
return err;
}
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
struct rt6_nh {
struct rt6_info *rt6_info;
struct fib6_config r_cfg;
struct mx6_config mxc;
struct list_head next;
};
static void ip6_print_replace_route_err(struct list_head *rt6_nh_list)
{
struct rt6_nh *nh;
list_for_each_entry(nh, rt6_nh_list, next) {
pr_warn("IPV6: multipath route replace failed (check consistency of installed routes): %pI6 nexthop %pI6 ifi %d\n",
&nh->r_cfg.fc_dst, &nh->r_cfg.fc_gateway,
nh->r_cfg.fc_ifindex);
}
}
static int ip6_route_info_append(struct list_head *rt6_nh_list,
struct rt6_info *rt, struct fib6_config *r_cfg)
{
struct rt6_nh *nh;
struct rt6_info *rtnh;
int err = -EEXIST;
list_for_each_entry(nh, rt6_nh_list, next) {
/* check if rt6_info already exists */
rtnh = nh->rt6_info;
if (rtnh->dst.dev == rt->dst.dev &&
rtnh->rt6i_idev == rt->rt6i_idev &&
ipv6_addr_equal(&rtnh->rt6i_gateway,
&rt->rt6i_gateway))
return err;
}
nh = kzalloc(sizeof(*nh), GFP_KERNEL);
if (!nh)
return -ENOMEM;
nh->rt6_info = rt;
err = ip6_convert_metrics(&nh->mxc, r_cfg);
if (err) {
kfree(nh);
return err;
}
memcpy(&nh->r_cfg, r_cfg, sizeof(*r_cfg));
list_add_tail(&nh->next, rt6_nh_list);
return 0;
}
static int ip6_route_multipath_add(struct fib6_config *cfg)
{
struct fib6_config r_cfg;
struct rtnexthop *rtnh;
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
struct rt6_info *rt;
struct rt6_nh *err_nh;
struct rt6_nh *nh, *nh_safe;
int remaining;
int attrlen;
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
int err = 1;
int nhn = 0;
int replace = (cfg->fc_nlinfo.nlh &&
(cfg->fc_nlinfo.nlh->nlmsg_flags & NLM_F_REPLACE));
LIST_HEAD(rt6_nh_list);
remaining = cfg->fc_mp_len;
rtnh = (struct rtnexthop *)cfg->fc_mp;
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
/* Parse a Multipath Entry and build a list (rt6_nh_list) of
* rt6_info structs per nexthop
*/
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) {
r_cfg.fc_gateway = nla_get_in6_addr(nla);
r_cfg.fc_flags |= RTF_GATEWAY;
}
r_cfg.fc_encap = nla_find(attrs, attrlen, RTA_ENCAP);
nla = nla_find(attrs, attrlen, RTA_ENCAP_TYPE);
if (nla)
r_cfg.fc_encap_type = nla_get_u16(nla);
}
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
err = ip6_route_info_create(&r_cfg, &rt);
if (err)
goto cleanup;
err = ip6_route_info_append(&rt6_nh_list, rt, &r_cfg);
if (err) {
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
dst_free(&rt->dst);
goto cleanup;
}
rtnh = rtnh_next(rtnh, &remaining);
}
err_nh = NULL;
list_for_each_entry(nh, &rt6_nh_list, next) {
err = __ip6_ins_rt(nh->rt6_info, &cfg->fc_nlinfo, &nh->mxc);
/* nh->rt6_info is used or freed at this point, reset to NULL*/
nh->rt6_info = NULL;
if (err) {
if (replace && nhn)
ip6_print_replace_route_err(&rt6_nh_list);
err_nh = nh;
goto add_errout;
}
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
/* Because each route is added like a single route we remove
* these flags after the first nexthop: if there is a collision,
* we have already failed to add the first nexthop:
* fib6_add_rt2node() has rejected it; when replacing, old
* nexthops have been replaced by first new, the rest should
* be added to it.
*/
cfg->fc_nlinfo.nlh->nlmsg_flags &= ~(NLM_F_EXCL |
NLM_F_REPLACE);
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
nhn++;
}
goto cleanup;
add_errout:
/* Delete routes that were already added */
list_for_each_entry(nh, &rt6_nh_list, next) {
if (err_nh == nh)
break;
ip6_route_del(&nh->r_cfg);
}
cleanup:
list_for_each_entry_safe(nh, nh_safe, &rt6_nh_list, next) {
if (nh->rt6_info)
dst_free(&nh->rt6_info->dst);
kfree(nh->mxc.mx);
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
list_del(&nh->next);
kfree(nh);
}
return err;
}
static int ip6_route_multipath_del(struct fib6_config *cfg)
{
struct fib6_config r_cfg;
struct rtnexthop *rtnh;
int remaining;
int attrlen;
int err = 1, last_err = 0;
remaining = cfg->fc_mp_len;
rtnh = (struct rtnexthop *)cfg->fc_mp;
/* 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 = ip6_route_del(&r_cfg);
if (err)
last_err = err;
rtnh = rtnh_next(rtnh, &remaining);
}
return last_err;
}
static int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh)
{
struct fib6_config cfg;
int err;
err = rtm_to_fib6_config(skb, nlh, &cfg);
if (err < 0)
return err;
if (cfg.fc_mp)
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
return ip6_route_multipath_del(&cfg);
else
return ip6_route_del(&cfg);
}
static int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh)
{
struct fib6_config cfg;
int err;
err = rtm_to_fib6_config(skb, nlh, &cfg);
if (err < 0)
return err;
if (cfg.fc_mp)
ipv6: fix multipath route replace error recovery Problem: The ecmp route replace support for ipv6 in the kernel, deletes the existing ecmp route too early, ie when it installs the first nexthop. If there is an error in installing the subsequent nexthops, its too late to recover the already deleted existing route leaving the fib in an inconsistent state. This patch reduces the possibility of this by doing the following: a) Changes the existing multipath route add code to a two stage process: build rt6_infos + insert them ip6_route_add rt6_info creation code is moved into ip6_route_info_create. b) This ensures that most errors are caught during building rt6_infos and we fail early c) Separates multipath add and del code. Because add needs the special two stage mode in a) and delete essentially does not care. d) In any event if the code fails during inserting a route again, a warning is printed (This should be unlikely) Before the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show /* previously added ecmp route 3000:1000:1000:1000::2 dissappears from * kernel */ After the patch: $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 /* Try replacing the route with a duplicate nexthop */ $ip -6 route change 3000:1000:1000:1000::2/128 nexthop via fe80::202:ff:fe00:b dev swp49s0 nexthop via fe80::202:ff:fe00:d dev swp49s1 nexthop via fe80::202:ff:fe00:d dev swp49s1 RTNETLINK answers: File exists $ip -6 route show 3000:1000:1000:1000::2 via fe80::202:ff:fe00:b dev swp49s0 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:d dev swp49s1 metric 1024 3000:1000:1000:1000::2 via fe80::202:ff:fe00:f dev swp49s2 metric 1024 Fixes: 27596472473a ("ipv6: fix ECMP route replacement") Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-08 17:53:04 +00:00
return ip6_route_multipath_add(&cfg);
else
return ip6_route_add(&cfg);
}
static inline size_t rt6_nlmsg_size(struct rt6_info *rt)
{
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 */
+ RTAX_MAX * nla_total_size(4) /* RTA_METRICS */
+ nla_total_size(sizeof(struct rta_cacheinfo))
+ nla_total_size(TCP_CA_NAME_MAX) /* RTAX_CC_ALGO */
+ nla_total_size(1) /* RTA_PREF */
+ lwtunnel_get_encap_size(rt->dst.lwtstate);
}
static int rt6_fill_node(struct net *net,
struct sk_buff *skb, struct rt6_info *rt,
struct in6_addr *dst, struct in6_addr *src,
int iif, int type, u32 portid, u32 seq,
int prefix, int nowait, unsigned int flags)
{
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
u32 metrics[RTAX_MAX];
struct rtmsg *rtm;
struct nlmsghdr *nlh;
long expires;
u32 table;
if (prefix) { /* user wants prefix routes only */
if (!(rt->rt6i_flags & RTF_PREFIX_RT)) {
/* success since this is not a prefix route */
return 1;
}
}
nlh = nlmsg_put(skb, portid, seq, type, sizeof(*rtm), flags);
if (!nlh)
return -EMSGSIZE;
rtm = nlmsg_data(nlh);
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;
if (rt->rt6i_table)
table = rt->rt6i_table->tb6_id;
else
table = RT6_TABLE_UNSPEC;
rtm->rtm_table = table;
if (nla_put_u32(skb, RTA_TABLE, table))
goto nla_put_failure;
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;
case -EAGAIN:
rtm->rtm_type = RTN_THROW;
break;
default:
rtm->rtm_type = RTN_UNREACHABLE;
break;
}
}
else if (rt->rt6i_flags & RTF_LOCAL)
rtm->rtm_type = RTN_LOCAL;
else if (rt->dst.dev && (rt->dst.dev->flags & IFF_LOOPBACK))
rtm->rtm_type = RTN_LOCAL;
else
rtm->rtm_type = RTN_UNICAST;
rtm->rtm_flags = 0;
net: ipv6 sysctl option to ignore routes when nexthop link is down Like the ipv4 patch with a similar title, this adds a sysctl to allow the user to change routing behavior based on whether or not the interface associated with the nexthop was an up or down link. The default setting preserves the current behavior, but anyone that enables it will notice that nexthops on down interfaces will no longer be selected: net.ipv6.conf.all.ignore_routes_with_linkdown = 0 net.ipv6.conf.default.ignore_routes_with_linkdown = 0 net.ipv6.conf.lo.ignore_routes_with_linkdown = 0 ... When the above sysctls are set, not only will link status be reported to userspace, but an indication that a nexthop is dead and will not be used is also reported. 1000::/8 via 7000::2 dev p7p1 metric 1024 dead linkdown pref medium 1000::/8 via 8000::2 dev p8p1 metric 1024 pref medium 7000::/8 dev p7p1 proto kernel metric 256 dead linkdown pref medium 8000::/8 dev p8p1 proto kernel metric 256 pref medium 9000::/8 via 8000::2 dev p8p1 metric 2048 pref medium 9000::/8 via 7000::2 dev p7p1 metric 1024 dead linkdown pref medium fe80::/64 dev p7p1 proto kernel metric 256 dead linkdown pref medium fe80::/64 dev p8p1 proto kernel metric 256 pref medium This also adds devconf support and notification when sysctl values change. v2: drop use of rt6i_nhflags since it is not needed right now Signed-off-by: Andy Gospodarek <gospo@cumulusnetworks.com> Signed-off-by: Dinesh Dutt <ddutt@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-13 14:39:01 +00:00
if (!netif_carrier_ok(rt->dst.dev)) {
rtm->rtm_flags |= RTNH_F_LINKDOWN;
net: ipv6 sysctl option to ignore routes when nexthop link is down Like the ipv4 patch with a similar title, this adds a sysctl to allow the user to change routing behavior based on whether or not the interface associated with the nexthop was an up or down link. The default setting preserves the current behavior, but anyone that enables it will notice that nexthops on down interfaces will no longer be selected: net.ipv6.conf.all.ignore_routes_with_linkdown = 0 net.ipv6.conf.default.ignore_routes_with_linkdown = 0 net.ipv6.conf.lo.ignore_routes_with_linkdown = 0 ... When the above sysctls are set, not only will link status be reported to userspace, but an indication that a nexthop is dead and will not be used is also reported. 1000::/8 via 7000::2 dev p7p1 metric 1024 dead linkdown pref medium 1000::/8 via 8000::2 dev p8p1 metric 1024 pref medium 7000::/8 dev p7p1 proto kernel metric 256 dead linkdown pref medium 8000::/8 dev p8p1 proto kernel metric 256 pref medium 9000::/8 via 8000::2 dev p8p1 metric 2048 pref medium 9000::/8 via 7000::2 dev p7p1 metric 1024 dead linkdown pref medium fe80::/64 dev p7p1 proto kernel metric 256 dead linkdown pref medium fe80::/64 dev p8p1 proto kernel metric 256 pref medium This also adds devconf support and notification when sysctl values change. v2: drop use of rt6i_nhflags since it is not needed right now Signed-off-by: Andy Gospodarek <gospo@cumulusnetworks.com> Signed-off-by: Dinesh Dutt <ddutt@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-13 14:39:01 +00:00
if (rt->rt6i_idev->cnf.ignore_routes_with_linkdown)
rtm->rtm_flags |= RTNH_F_DEAD;
}
rtm->rtm_scope = RT_SCOPE_UNIVERSE;
rtm->rtm_protocol = rt->rt6i_protocol;
if (rt->rt6i_flags & RTF_DYNAMIC)
rtm->rtm_protocol = RTPROT_REDIRECT;
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;
}
if (rt->rt6i_flags & RTF_CACHE)
rtm->rtm_flags |= RTM_F_CLONED;
if (dst) {
if (nla_put_in6_addr(skb, RTA_DST, dst))
goto nla_put_failure;
rtm->rtm_dst_len = 128;
} else if (rtm->rtm_dst_len)
if (nla_put_in6_addr(skb, RTA_DST, &rt->rt6i_dst.addr))
goto nla_put_failure;
#ifdef CONFIG_IPV6_SUBTREES
if (src) {
if (nla_put_in6_addr(skb, RTA_SRC, src))
goto nla_put_failure;
rtm->rtm_src_len = 128;
} else if (rtm->rtm_src_len &&
nla_put_in6_addr(skb, RTA_SRC, &rt->rt6i_src.addr))
goto nla_put_failure;
#endif
if (iif) {
#ifdef CONFIG_IPV6_MROUTE
if (ipv6_addr_is_multicast(&rt->rt6i_dst.addr)) {
int err = ip6mr_get_route(net, skb, rtm, nowait);
if (err <= 0) {
if (!nowait) {
if (err == 0)
return 0;
goto nla_put_failure;
} else {
if (err == -EMSGSIZE)
goto nla_put_failure;
}
}
} else
#endif
if (nla_put_u32(skb, RTA_IIF, iif))
goto nla_put_failure;
} else if (dst) {
struct in6_addr saddr_buf;
if (ip6_route_get_saddr(net, rt, dst, 0, &saddr_buf) == 0 &&
nla_put_in6_addr(skb, RTA_PREFSRC, &saddr_buf))
goto nla_put_failure;
}
if (rt->rt6i_prefsrc.plen) {
struct in6_addr saddr_buf;
saddr_buf = rt->rt6i_prefsrc.addr;
if (nla_put_in6_addr(skb, RTA_PREFSRC, &saddr_buf))
goto nla_put_failure;
}
ipv6: Stop rt6_info from using inet_peer's metrics inet_peer is indexed by the dst address alone. However, the fib6 tree could have multiple routing entries (rt6_info) for the same dst. For example, 1. A /128 dst via multiple gateways. 2. A RTF_CACHE route cloned from a /128 route. In the above cases, all of them will share the same metrics and step on each other. This patch will steer away from inet_peer's metrics and use dst_cow_metrics_generic() for everything. Change Highlights: 1. Remove rt6_cow_metrics() which currently acquires metrics from inet_peer for DST_HOST route (i.e. /128 route). 2. Add rt6i_pmtu to take care of the pmtu update to avoid creating a full size metrics just to override the RTAX_MTU. 3. After (2), the RTF_CACHE route can also share the metrics with its dst.from route, by: dst_init_metrics(&cache_rt->dst, dst_metrics_ptr(cache_rt->dst.from), true); 4. Stop creating RTF_CACHE route by cloning another RTF_CACHE route. Instead, directly clone from rt->dst. [ Currently, cloning from another RTF_CACHE is only possible during rt6_do_redirect(). Also, the old clone is removed from the tree immediately after the new clone is added. ] In case of cloning from an older redirect RTF_CACHE, it should work as before. In case of cloning from an older pmtu RTF_CACHE, this patch will forget the pmtu and re-learn it (if there is any) from the redirected route. The _rt6i_peer and DST_METRICS_FORCE_OVERWRITE will be removed in the next cleanup patch. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-28 20:03:06 +00:00
memcpy(metrics, dst_metrics_ptr(&rt->dst), sizeof(metrics));
if (rt->rt6i_pmtu)
metrics[RTAX_MTU - 1] = rt->rt6i_pmtu;
if (rtnetlink_put_metrics(skb, metrics) < 0)
goto nla_put_failure;
if (rt->rt6i_flags & RTF_GATEWAY) {
if (nla_put_in6_addr(skb, RTA_GATEWAY, &rt->rt6i_gateway) < 0)
goto nla_put_failure;
}
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;
expires = (rt->rt6i_flags & RTF_EXPIRES) ? rt->dst.expires - jiffies : 0;
if (rtnl_put_cacheinfo(skb, &rt->dst, 0, expires, rt->dst.error) < 0)
goto nla_put_failure;
if (nla_put_u8(skb, RTA_PREF, IPV6_EXTRACT_PREF(rt->rt6i_flags)))
goto nla_put_failure;
lwtunnel_fill_encap(skb, rt->dst.lwtstate);
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
nlmsg_end(skb, nlh);
return 0;
nla_put_failure:
nlmsg_cancel(skb, nlh);
return -EMSGSIZE;
}
int rt6_dump_route(struct rt6_info *rt, void *p_arg)
{
struct rt6_rtnl_dump_arg *arg = (struct rt6_rtnl_dump_arg *) p_arg;
int prefix;
if (nlmsg_len(arg->cb->nlh) >= sizeof(struct rtmsg)) {
struct rtmsg *rtm = nlmsg_data(arg->cb->nlh);
prefix = (rtm->rtm_flags & RTM_F_PREFIX) != 0;
} else
prefix = 0;
return rt6_fill_node(arg->net,
arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE,
NETLINK_CB(arg->cb->skb).portid, arg->cb->nlh->nlmsg_seq,
prefix, 0, NLM_F_MULTI);
}
static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh)
{
struct net *net = sock_net(in_skb->sk);
struct nlattr *tb[RTA_MAX+1];
struct rt6_info *rt;
struct sk_buff *skb;
struct rtmsg *rtm;
struct flowi6 fl6;
int err, iif = 0, oif = 0;
err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv6_policy);
if (err < 0)
goto errout;
err = -EINVAL;
memset(&fl6, 0, sizeof(fl6));
if (tb[RTA_SRC]) {
if (nla_len(tb[RTA_SRC]) < sizeof(struct in6_addr))
goto errout;
fl6.saddr = *(struct in6_addr *)nla_data(tb[RTA_SRC]);
}
if (tb[RTA_DST]) {
if (nla_len(tb[RTA_DST]) < sizeof(struct in6_addr))
goto errout;
fl6.daddr = *(struct in6_addr *)nla_data(tb[RTA_DST]);
}
if (tb[RTA_IIF])
iif = nla_get_u32(tb[RTA_IIF]);
if (tb[RTA_OIF])
oif = nla_get_u32(tb[RTA_OIF]);
if (tb[RTA_MARK])
fl6.flowi6_mark = nla_get_u32(tb[RTA_MARK]);
if (iif) {
struct net_device *dev;
int flags = 0;
dev = __dev_get_by_index(net, iif);
if (!dev) {
err = -ENODEV;
goto errout;
}
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);
}
skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
if (!skb) {
ip6_rt_put(rt);
err = -ENOBUFS;
goto errout;
}
/* Reserve room for dummy headers, this skb can pass
through good chunk of routing engine.
*/
skb_reset_mac_header(skb);
skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr));
skb_dst_set(skb, &rt->dst);
err = rt6_fill_node(net, skb, rt, &fl6.daddr, &fl6.saddr, iif,
RTM_NEWROUTE, NETLINK_CB(in_skb).portid,
nlh->nlmsg_seq, 0, 0, 0);
if (err < 0) {
kfree_skb(skb);
goto errout;
}
err = rtnl_unicast(skb, net, NETLINK_CB(in_skb).portid);
errout:
return err;
}
void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info,
unsigned int nlm_flags)
{
struct sk_buff *skb;
struct net *net = info->nl_net;
u32 seq;
int err;
err = -ENOBUFS;
seq = info->nlh ? info->nlh->nlmsg_seq : 0;
skb = nlmsg_new(rt6_nlmsg_size(rt), gfp_any());
if (!skb)
goto errout;
err = rt6_fill_node(net, skb, rt, NULL, NULL, 0,
event, info->portid, seq, 0, 0, nlm_flags);
if (err < 0) {
/* -EMSGSIZE implies BUG in rt6_nlmsg_size() */
WARN_ON(err == -EMSGSIZE);
kfree_skb(skb);
goto errout;
}
rtnl_notify(skb, net, info->portid, RTNLGRP_IPV6_ROUTE,
2009-02-25 07:18:28 +00:00
info->nlh, gfp_any());
return;
errout:
if (err < 0)
rtnl_set_sk_err(net, RTNLGRP_IPV6_ROUTE, err);
}
static int ip6_route_dev_notify(struct notifier_block *this,
unsigned long event, void *ptr)
{
struct net_device *dev = netdev_notifier_info_to_dev(ptr);
struct net *net = dev_net(dev);
if (event == NETDEV_REGISTER && (dev->flags & IFF_LOOPBACK)) {
net->ipv6.ip6_null_entry->dst.dev = dev;
net->ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(dev);
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
net->ipv6.ip6_prohibit_entry->dst.dev = dev;
net->ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(dev);
net->ipv6.ip6_blk_hole_entry->dst.dev = dev;
net->ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(dev);
#endif
}
return NOTIFY_OK;
}
/*
* /proc
*/
#ifdef CONFIG_PROC_FS
static const struct file_operations ipv6_route_proc_fops = {
.owner = THIS_MODULE,
.open = ipv6_route_open,
.read = seq_read,
.llseek = seq_lseek,
.release = seq_release_net,
};
static int rt6_stats_seq_show(struct seq_file *seq, void *v)
{
struct net *net = (struct net *)seq->private;
seq_printf(seq, "%04x %04x %04x %04x %04x %04x %04x\n",
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,
dst_entries_get_slow(&net->ipv6.ip6_dst_ops),
net->ipv6.rt6_stats->fib_discarded_routes);
return 0;
}
static int rt6_stats_seq_open(struct inode *inode, struct file *file)
{
return single_open_net(inode, file, rt6_stats_seq_show);
}
static const struct file_operations rt6_stats_seq_fops = {
.owner = THIS_MODULE,
.open = rt6_stats_seq_open,
.read = seq_read,
.llseek = seq_lseek,
.release = single_release_net,
};
#endif /* CONFIG_PROC_FS */
#ifdef CONFIG_SYSCTL
static
int ipv6_sysctl_rtcache_flush(struct ctl_table *ctl, int write,
void __user *buffer, size_t *lenp, loff_t *ppos)
{
struct net *net;
int delay;
if (!write)
return -EINVAL;
net = (struct net *)ctl->extra1;
delay = net->ipv6.sysctl.flush_delay;
proc_dointvec(ctl, write, buffer, lenp, ppos);
fib6_run_gc(delay <= 0 ? 0 : (unsigned long)delay, net, delay > 0);
return 0;
}
struct ctl_table ipv6_route_table_template[] = {
{
.procname = "flush",
.data = &init_net.ipv6.sysctl.flush_delay,
.maxlen = sizeof(int),
.mode = 0200,
.proc_handler = ipv6_sysctl_rtcache_flush
},
{
.procname = "gc_thresh",
.data = &ip6_dst_ops_template.gc_thresh,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec,
},
{
.procname = "max_size",
.data = &init_net.ipv6.sysctl.ip6_rt_max_size,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec,
},
{
.procname = "gc_min_interval",
.data = &init_net.ipv6.sysctl.ip6_rt_gc_min_interval,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_jiffies,
},
{
.procname = "gc_timeout",
.data = &init_net.ipv6.sysctl.ip6_rt_gc_timeout,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_jiffies,
},
{
.procname = "gc_interval",
.data = &init_net.ipv6.sysctl.ip6_rt_gc_interval,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_jiffies,
},
{
.procname = "gc_elasticity",
.data = &init_net.ipv6.sysctl.ip6_rt_gc_elasticity,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec,
},
{
.procname = "mtu_expires",
.data = &init_net.ipv6.sysctl.ip6_rt_mtu_expires,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_jiffies,
},
{
.procname = "min_adv_mss",
.data = &init_net.ipv6.sysctl.ip6_rt_min_advmss,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec,
},
{
.procname = "gc_min_interval_ms",
.data = &init_net.ipv6.sysctl.ip6_rt_gc_min_interval,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_ms_jiffies,
},
{ }
};
struct ctl_table * __net_init ipv6_route_sysctl_init(struct net *net)
{
struct ctl_table *table;
table = kmemdup(ipv6_route_table_template,
sizeof(ipv6_route_table_template),
GFP_KERNEL);
if (table) {
table[0].data = &net->ipv6.sysctl.flush_delay;
table[0].extra1 = net;
table[1].data = &net->ipv6.ip6_dst_ops.gc_thresh;
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;
table[9].data = &net->ipv6.sysctl.ip6_rt_gc_min_interval;
/* Don't export sysctls to unprivileged users */
if (net->user_ns != &init_user_ns)
table[0].procname = NULL;
}
return table;
}
#endif
static int __net_init ip6_route_net_init(struct net *net)
{
int ret = -ENOMEM;
memcpy(&net->ipv6.ip6_dst_ops, &ip6_dst_ops_template,
sizeof(net->ipv6.ip6_dst_ops));
if (dst_entries_init(&net->ipv6.ip6_dst_ops) < 0)
goto out_ip6_dst_ops;
net->ipv6.ip6_null_entry = kmemdup(&ip6_null_entry_template,
sizeof(*net->ipv6.ip6_null_entry),
GFP_KERNEL);
if (!net->ipv6.ip6_null_entry)
goto out_ip6_dst_entries;
net->ipv6.ip6_null_entry->dst.path =
(struct dst_entry *)net->ipv6.ip6_null_entry;
net->ipv6.ip6_null_entry->dst.ops = &net->ipv6.ip6_dst_ops;
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>
2011-01-27 04:51:05 +00:00
dst_init_metrics(&net->ipv6.ip6_null_entry->dst,
ip6_template_metrics, true);
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
net->ipv6.ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template,
sizeof(*net->ipv6.ip6_prohibit_entry),
GFP_KERNEL);
if (!net->ipv6.ip6_prohibit_entry)
goto out_ip6_null_entry;
net->ipv6.ip6_prohibit_entry->dst.path =
(struct dst_entry *)net->ipv6.ip6_prohibit_entry;
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>
2011-01-27 04:51:05 +00:00
dst_init_metrics(&net->ipv6.ip6_prohibit_entry->dst,
ip6_template_metrics, true);
net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template,
sizeof(*net->ipv6.ip6_blk_hole_entry),
GFP_KERNEL);
if (!net->ipv6.ip6_blk_hole_entry)
goto out_ip6_prohibit_entry;
net->ipv6.ip6_blk_hole_entry->dst.path =
(struct dst_entry *)net->ipv6.ip6_blk_hole_entry;
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>
2011-01-27 04:51:05 +00:00
dst_init_metrics(&net->ipv6.ip6_blk_hole_entry->dst,
ip6_template_metrics, true);
#endif
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;
net->ipv6.ip6_rt_gc_expire = 30*HZ;
ret = 0;
out:
return ret;
#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
out_ip6_dst_entries:
dst_entries_destroy(&net->ipv6.ip6_dst_ops);
out_ip6_dst_ops:
goto out;
}
static void __net_exit ip6_route_net_exit(struct net *net)
{
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
dst_entries_destroy(&net->ipv6.ip6_dst_ops);
}
static int __net_init ip6_route_net_init_late(struct net *net)
{
#ifdef CONFIG_PROC_FS
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);
#endif
return 0;
}
static void __net_exit ip6_route_net_exit_late(struct net *net)
{
#ifdef CONFIG_PROC_FS
remove_proc_entry("ipv6_route", net->proc_net);
remove_proc_entry("rt6_stats", net->proc_net);
#endif
}
static struct pernet_operations ip6_route_net_ops = {
.init = ip6_route_net_init,
.exit = ip6_route_net_exit,
};
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;
inetpeer_invalidate_tree(bp);
kfree(bp);
}
static struct pernet_operations ipv6_inetpeer_ops = {
.init = ipv6_inetpeer_init,
.exit = ipv6_inetpeer_exit,
};
static struct pernet_operations ip6_route_net_late_ops = {
.init = ip6_route_net_init_late,
.exit = ip6_route_net_exit_late,
};
static struct notifier_block ip6_route_dev_notifier = {
.notifier_call = ip6_route_dev_notify,
.priority = 0,
};
int __init ip6_route_init(void)
{
int ret;
int cpu;
ret = -ENOMEM;
ip6_dst_ops_template.kmem_cachep =
kmem_cache_create("ip6_dst_cache", sizeof(struct rt6_info), 0,
SLAB_HWCACHE_ALIGN, NULL);
if (!ip6_dst_ops_template.kmem_cachep)
goto out;
ret = dst_entries_init(&ip6_dst_blackhole_ops);
if (ret)
goto out_kmem_cache;
ret = register_pernet_subsys(&ipv6_inetpeer_ops);
if (ret)
goto out_dst_entries;
ret = register_pernet_subsys(&ip6_route_net_ops);
if (ret)
goto out_register_inetpeer;
XFRM,IPv6: initialize ip6_dst_blackhole_ops.kmem_cachep ip6_dst_blackhole_ops.kmem_cachep is not expected to be NULL (i.e. to be initialized) when dst_alloc() is called from ip6_dst_blackhole(). Otherwise, it results in the following (xfrm_larval_drop is now set to 1 by default): [ 78.697642] Unable to handle kernel paging request for data at address 0x0000004c [ 78.703449] Faulting instruction address: 0xc0097f54 [ 78.786896] Oops: Kernel access of bad area, sig: 11 [#1] [ 78.792791] PowerMac [ 78.798383] Modules linked in: btusb usbhid bluetooth b43 mac80211 cfg80211 ehci_hcd ohci_hcd sungem sungem_phy usbcore ssb [ 78.804263] NIP: c0097f54 LR: c0334a28 CTR: c002d430 [ 78.809997] REGS: eef19ad0 TRAP: 0300 Not tainted (2.6.27-rc5) [ 78.815743] MSR: 00001032 <ME,IR,DR> CR: 22242482 XER: 20000000 [ 78.821550] DAR: 0000004c, DSISR: 40000000 [ 78.827278] TASK = eef0df40[3035] 'mip6d' THREAD: eef18000 [ 78.827408] GPR00: 00001032 eef19b80 eef0df40 00000000 00008020 eef19c30 00000001 00000000 [ 78.833249] GPR08: eee5101c c05a5c10 ef9ad500 00000000 24242422 1005787c 00000000 1004f960 [ 78.839151] GPR16: 00000000 10024e90 10050040 48030018 0fe44150 00000000 00000000 eef19c30 [ 78.845046] GPR24: eef19e44 00000000 eef19bf8 efb37c14 eef19bf8 00008020 00009032 c0596064 [ 78.856671] NIP [c0097f54] kmem_cache_alloc+0x20/0x94 [ 78.862581] LR [c0334a28] dst_alloc+0x40/0xc4 [ 78.868451] Call Trace: [ 78.874252] [eef19b80] [c03c1810] ip6_dst_lookup_tail+0x1c8/0x1dc (unreliable) [ 78.880222] [eef19ba0] [c0334a28] dst_alloc+0x40/0xc4 [ 78.886164] [eef19bb0] [c03cd698] ip6_dst_blackhole+0x28/0x1cc [ 78.892090] [eef19be0] [c03d9be8] rawv6_sendmsg+0x75c/0xc88 [ 78.897999] [eef19cb0] [c038bca4] inet_sendmsg+0x4c/0x78 [ 78.903907] [eef19cd0] [c03207c8] sock_sendmsg+0xac/0xe4 [ 78.909734] [eef19db0] [c03209e4] sys_sendmsg+0x1e4/0x2a0 [ 78.915540] [eef19f00] [c03220a8] sys_socketcall+0xfc/0x210 [ 78.921406] [eef19f40] [c0014b3c] ret_from_syscall+0x0/0x38 [ 78.927295] --- Exception: c01 at 0xfe2d730 [ 78.927297] LR = 0xfe2d71c [ 78.939019] Instruction dump: [ 78.944835] 91640018 9144001c 900a0000 4bffff44 9421ffe0 7c0802a6 bf810010 7c9d2378 [ 78.950694] 90010024 7fc000a6 57c0045e 7c000124 <83e3004c> 8383005c 2f9f0000 419e0050 [ 78.956464] ---[ end trace 05fa1ed7972487a1 ]--- As commented by Benjamin Thery, the bug was introduced by f2fc6a54585a1be6669613a31fbaba2ecbadcd36, while adding network namespaces support to ipv6 routes. Signed-off-by: Arnaud Ebalard <arno@natisbad.org> Acked-by: Benjamin Thery <benjamin.thery@bull.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-01 09:37:56 +00:00
ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops_template.kmem_cachep;
/* 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 */
init_net.ipv6.ip6_null_entry->dst.dev = init_net.loopback_dev;
init_net.ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
init_net.ipv6.ip6_prohibit_entry->dst.dev = init_net.loopback_dev;
init_net.ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
init_net.ipv6.ip6_blk_hole_entry->dst.dev = init_net.loopback_dev;
init_net.ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
#endif
ret = fib6_init();
if (ret)
goto out_register_subsys;
ret = xfrm6_init();
if (ret)
goto out_fib6_init;
ret = fib6_rules_init();
if (ret)
goto xfrm6_init;
ret = register_pernet_subsys(&ip6_route_net_late_ops);
if (ret)
goto fib6_rules_init;
ret = -ENOBUFS;
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))
goto out_register_late_subsys;
ret = register_netdevice_notifier(&ip6_route_dev_notifier);
if (ret)
goto out_register_late_subsys;
for_each_possible_cpu(cpu) {
struct uncached_list *ul = per_cpu_ptr(&rt6_uncached_list, cpu);
INIT_LIST_HEAD(&ul->head);
spin_lock_init(&ul->lock);
}
out:
return ret;
out_register_late_subsys:
unregister_pernet_subsys(&ip6_route_net_late_ops);
fib6_rules_init:
fib6_rules_cleanup();
xfrm6_init:
xfrm6_fini();
out_fib6_init:
fib6_gc_cleanup();
out_register_subsys:
unregister_pernet_subsys(&ip6_route_net_ops);
out_register_inetpeer:
unregister_pernet_subsys(&ipv6_inetpeer_ops);
out_dst_entries:
dst_entries_destroy(&ip6_dst_blackhole_ops);
out_kmem_cache:
kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep);
goto out;
}
void ip6_route_cleanup(void)
{
unregister_netdevice_notifier(&ip6_route_dev_notifier);
unregister_pernet_subsys(&ip6_route_net_late_ops);
fib6_rules_cleanup();
xfrm6_fini();
fib6_gc_cleanup();
unregister_pernet_subsys(&ipv6_inetpeer_ops);
unregister_pernet_subsys(&ip6_route_net_ops);
dst_entries_destroy(&ip6_dst_blackhole_ops);
kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep);
}