2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* INET An implementation of the TCP/IP protocol suite for the LINUX
|
|
|
|
* operating system. INET is implemented using the BSD Socket
|
|
|
|
* interface as the means of communication with the user level.
|
|
|
|
*
|
|
|
|
* IPv4 Forwarding Information Base: semantics.
|
|
|
|
*
|
|
|
|
* Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
|
|
|
|
*
|
|
|
|
* 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.
|
|
|
|
*/
|
|
|
|
|
2016-12-24 12:46:01 -07:00
|
|
|
#include <linux/uaccess.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/bitops.h>
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/jiffies.h>
|
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/socket.h>
|
|
|
|
#include <linux/sockios.h>
|
|
|
|
#include <linux/errno.h>
|
|
|
|
#include <linux/in.h>
|
|
|
|
#include <linux/inet.h>
|
2005-12-26 21:43:12 -07:00
|
|
|
#include <linux/inetdevice.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/netdevice.h>
|
|
|
|
#include <linux/if_arp.h>
|
|
|
|
#include <linux/proc_fs.h>
|
|
|
|
#include <linux/skbuff.h>
|
|
|
|
#include <linux/init.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 02:04:11 -06:00
|
|
|
#include <linux/slab.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2005-12-26 21:43:12 -07:00
|
|
|
#include <net/arp.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <net/ip.h>
|
|
|
|
#include <net/protocol.h>
|
|
|
|
#include <net/route.h>
|
|
|
|
#include <net/tcp.h>
|
|
|
|
#include <net/sock.h>
|
|
|
|
#include <net/ip_fib.h>
|
2006-08-15 01:34:17 -06:00
|
|
|
#include <net/netlink.h>
|
2006-08-17 19:14:52 -06:00
|
|
|
#include <net/nexthop.h>
|
2015-07-21 02:43:47 -06:00
|
|
|
#include <net/lwtunnel.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#include "fib_lookup.h"
|
|
|
|
|
2006-08-29 17:48:09 -06:00
|
|
|
static DEFINE_SPINLOCK(fib_info_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
static struct hlist_head *fib_info_hash;
|
|
|
|
static struct hlist_head *fib_info_laddrhash;
|
2011-02-01 16:34:21 -07:00
|
|
|
static unsigned int fib_info_hash_size;
|
2005-04-16 16:20:36 -06:00
|
|
|
static unsigned int fib_info_cnt;
|
|
|
|
|
|
|
|
#define DEVINDEX_HASHBITS 8
|
|
|
|
#define DEVINDEX_HASHSIZE (1U << DEVINDEX_HASHBITS)
|
|
|
|
static struct hlist_head fib_info_devhash[DEVINDEX_HASHSIZE];
|
|
|
|
|
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
2015-09-30 02:12:21 -06:00
|
|
|
u32 fib_multipath_secret __read_mostly;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-10-04 14:00:18 -06:00
|
|
|
#define for_nexthops(fi) { \
|
|
|
|
int nhsel; const struct fib_nh *nh; \
|
|
|
|
for (nhsel = 0, nh = (fi)->fib_nh; \
|
|
|
|
nhsel < (fi)->fib_nhs; \
|
|
|
|
nh++, nhsel++)
|
|
|
|
|
|
|
|
#define change_nexthops(fi) { \
|
|
|
|
int nhsel; struct fib_nh *nexthop_nh; \
|
|
|
|
for (nhsel = 0, nexthop_nh = (struct fib_nh *)((fi)->fib_nh); \
|
|
|
|
nhsel < (fi)->fib_nhs; \
|
|
|
|
nexthop_nh++, nhsel++)
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#else /* CONFIG_IP_ROUTE_MULTIPATH */
|
|
|
|
|
|
|
|
/* Hope, that gcc will optimize it to get rid of dummy loop */
|
|
|
|
|
2010-10-04 14:00:18 -06:00
|
|
|
#define for_nexthops(fi) { \
|
|
|
|
int nhsel; const struct fib_nh *nh = (fi)->fib_nh; \
|
|
|
|
for (nhsel = 0; nhsel < 1; nhsel++)
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-10-04 14:00:18 -06:00
|
|
|
#define change_nexthops(fi) { \
|
|
|
|
int nhsel; \
|
|
|
|
struct fib_nh *nexthop_nh = (struct fib_nh *)((fi)->fib_nh); \
|
|
|
|
for (nhsel = 0; nhsel < 1; nhsel++)
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#endif /* CONFIG_IP_ROUTE_MULTIPATH */
|
|
|
|
|
|
|
|
#define endfor_nexthops(fi) }
|
|
|
|
|
|
|
|
|
2011-03-07 16:01:10 -07:00
|
|
|
const struct fib_prop fib_props[RTN_MAX + 1] = {
|
2010-10-04 14:00:18 -06:00
|
|
|
[RTN_UNSPEC] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = 0,
|
|
|
|
.scope = RT_SCOPE_NOWHERE,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_UNICAST] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = 0,
|
|
|
|
.scope = RT_SCOPE_UNIVERSE,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_LOCAL] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = 0,
|
|
|
|
.scope = RT_SCOPE_HOST,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_BROADCAST] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = 0,
|
|
|
|
.scope = RT_SCOPE_LINK,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_ANYCAST] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = 0,
|
|
|
|
.scope = RT_SCOPE_LINK,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_MULTICAST] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = 0,
|
|
|
|
.scope = RT_SCOPE_UNIVERSE,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_BLACKHOLE] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = -EINVAL,
|
|
|
|
.scope = RT_SCOPE_UNIVERSE,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_UNREACHABLE] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = -EHOSTUNREACH,
|
|
|
|
.scope = RT_SCOPE_UNIVERSE,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_PROHIBIT] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = -EACCES,
|
|
|
|
.scope = RT_SCOPE_UNIVERSE,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_THROW] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = -EAGAIN,
|
|
|
|
.scope = RT_SCOPE_UNIVERSE,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_NAT] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = -EINVAL,
|
|
|
|
.scope = RT_SCOPE_NOWHERE,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
|
|
|
[RTN_XRESOLVE] = {
|
2005-04-16 16:20:36 -06:00
|
|
|
.error = -EINVAL,
|
|
|
|
.scope = RT_SCOPE_NOWHERE,
|
2010-10-04 14:00:18 -06:00
|
|
|
},
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2012-07-31 16:02:02 -06:00
|
|
|
static void rt_fibinfo_free(struct rtable __rcu **rtp)
|
|
|
|
{
|
|
|
|
struct rtable *rt = rcu_dereference_protected(*rtp, 1);
|
|
|
|
|
|
|
|
if (!rt)
|
|
|
|
return;
|
|
|
|
|
|
|
|
/* Not even needed : RCU_INIT_POINTER(*rtp, NULL);
|
|
|
|
* because we waited an RCU grace period before calling
|
|
|
|
* free_fib_info_rcu()
|
|
|
|
*/
|
|
|
|
|
|
|
|
dst_free(&rt->dst);
|
|
|
|
}
|
|
|
|
|
2012-07-17 05:19:00 -06:00
|
|
|
static void free_nh_exceptions(struct fib_nh *nh)
|
|
|
|
{
|
2014-09-03 23:21:56 -06:00
|
|
|
struct fnhe_hash_bucket *hash;
|
2012-07-17 05:19:00 -06:00
|
|
|
int i;
|
|
|
|
|
2014-09-03 23:21:56 -06:00
|
|
|
hash = rcu_dereference_protected(nh->nh_exceptions, 1);
|
|
|
|
if (!hash)
|
|
|
|
return;
|
2012-07-17 05:19:00 -06:00
|
|
|
for (i = 0; i < FNHE_HASH_SIZE; i++) {
|
|
|
|
struct fib_nh_exception *fnhe;
|
|
|
|
|
2012-07-17 14:42:13 -06:00
|
|
|
fnhe = rcu_dereference_protected(hash[i].chain, 1);
|
2012-07-17 05:19:00 -06:00
|
|
|
while (fnhe) {
|
|
|
|
struct fib_nh_exception *next;
|
|
|
|
|
2012-07-17 14:42:13 -06:00
|
|
|
next = rcu_dereference_protected(fnhe->fnhe_next, 1);
|
2012-07-31 16:02:02 -06:00
|
|
|
|
2013-06-27 01:27:05 -06:00
|
|
|
rt_fibinfo_free(&fnhe->fnhe_rth_input);
|
|
|
|
rt_fibinfo_free(&fnhe->fnhe_rth_output);
|
2012-07-31 16:02:02 -06:00
|
|
|
|
2012-07-17 05:19:00 -06:00
|
|
|
kfree(fnhe);
|
|
|
|
|
|
|
|
fnhe = next;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
kfree(hash);
|
|
|
|
}
|
|
|
|
|
2012-07-31 16:02:02 -06:00
|
|
|
static void rt_fibinfo_free_cpus(struct rtable __rcu * __percpu *rtp)
|
2012-07-30 23:45:30 -06:00
|
|
|
{
|
|
|
|
int cpu;
|
|
|
|
|
|
|
|
if (!rtp)
|
|
|
|
return;
|
|
|
|
|
|
|
|
for_each_possible_cpu(cpu) {
|
|
|
|
struct rtable *rt;
|
|
|
|
|
|
|
|
rt = rcu_dereference_protected(*per_cpu_ptr(rtp, cpu), 1);
|
|
|
|
if (rt)
|
|
|
|
dst_free(&rt->dst);
|
|
|
|
}
|
|
|
|
free_percpu(rtp);
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/* Release a nexthop info record */
|
2011-09-04 14:24:20 -06:00
|
|
|
static void free_fib_info_rcu(struct rcu_head *head)
|
|
|
|
{
|
|
|
|
struct fib_info *fi = container_of(head, struct fib_info, rcu);
|
|
|
|
|
2012-05-23 09:39:45 -06:00
|
|
|
change_nexthops(fi) {
|
|
|
|
if (nexthop_nh->nh_dev)
|
|
|
|
dev_put(nexthop_nh->nh_dev);
|
2015-07-24 04:28:36 -06:00
|
|
|
lwtstate_put(nexthop_nh->nh_lwtstate);
|
2014-09-03 23:21:56 -06:00
|
|
|
free_nh_exceptions(nexthop_nh);
|
2012-07-31 16:02:02 -06:00
|
|
|
rt_fibinfo_free_cpus(nexthop_nh->nh_pcpu_rth_output);
|
|
|
|
rt_fibinfo_free(&nexthop_nh->nh_rth_input);
|
2012-05-23 09:39:45 -06:00
|
|
|
} endfor_nexthops(fi);
|
|
|
|
|
2011-09-04 14:24:20 -06:00
|
|
|
if (fi->fib_metrics != (u32 *) dst_default_metrics)
|
|
|
|
kfree(fi->fib_metrics);
|
|
|
|
kfree(fi);
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
void free_fib_info(struct fib_info *fi)
|
|
|
|
{
|
|
|
|
if (fi->fib_dead == 0) {
|
2012-03-11 12:36:11 -06:00
|
|
|
pr_warn("Freeing alive fib_info %p\n", fi);
|
2005-04-16 16:20:36 -06:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
fib_info_cnt--;
|
2012-06-29 02:32:45 -06:00
|
|
|
#ifdef CONFIG_IP_ROUTE_CLASSID
|
|
|
|
change_nexthops(fi) {
|
|
|
|
if (nexthop_nh->nh_tclassid)
|
2012-07-05 23:13:13 -06:00
|
|
|
fi->fib_net->ipv4.fib_num_tclassid_users--;
|
2012-06-29 02:32:45 -06:00
|
|
|
} endfor_nexthops(fi);
|
|
|
|
#endif
|
2011-09-04 14:24:20 -06:00
|
|
|
call_rcu(&fi->rcu, free_fib_info_rcu);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2016-12-03 08:44:58 -07:00
|
|
|
EXPORT_SYMBOL_GPL(free_fib_info);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
void fib_release_info(struct fib_info *fi)
|
|
|
|
{
|
2006-08-29 17:48:09 -06:00
|
|
|
spin_lock_bh(&fib_info_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (fi && --fi->fib_treeref == 0) {
|
|
|
|
hlist_del(&fi->fib_hash);
|
|
|
|
if (fi->fib_prefsrc)
|
|
|
|
hlist_del(&fi->fib_lhash);
|
|
|
|
change_nexthops(fi) {
|
2010-01-15 02:16:40 -07:00
|
|
|
if (!nexthop_nh->nh_dev)
|
2005-04-16 16:20:36 -06:00
|
|
|
continue;
|
2010-01-15 02:16:40 -07:00
|
|
|
hlist_del(&nexthop_nh->nh_hash);
|
2005-04-16 16:20:36 -06:00
|
|
|
} endfor_nexthops(fi)
|
|
|
|
fi->fib_dead = 1;
|
|
|
|
fib_info_put(fi);
|
|
|
|
}
|
2006-08-29 17:48:09 -06:00
|
|
|
spin_unlock_bh(&fib_info_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2010-10-04 14:00:18 -06:00
|
|
|
static inline int nh_comp(const struct fib_info *fi, const struct fib_info *ofi)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
const struct fib_nh *onh = ofi->fib_nh;
|
|
|
|
|
|
|
|
for_nexthops(fi) {
|
|
|
|
if (nh->nh_oif != onh->nh_oif ||
|
|
|
|
nh->nh_gw != onh->nh_gw ||
|
|
|
|
nh->nh_scope != onh->nh_scope ||
|
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
|
|
|
nh->nh_weight != onh->nh_weight ||
|
|
|
|
#endif
|
2011-01-14 05:36:42 -07:00
|
|
|
#ifdef CONFIG_IP_ROUTE_CLASSID
|
2005-04-16 16:20:36 -06:00
|
|
|
nh->nh_tclassid != onh->nh_tclassid ||
|
|
|
|
#endif
|
2015-07-21 02:43:47 -06:00
|
|
|
lwtunnel_cmp_encap(nh->nh_lwtstate, onh->nh_lwtstate) ||
|
2015-06-23 11:45:36 -06:00
|
|
|
((nh->nh_flags ^ onh->nh_flags) & ~RTNH_COMPARE_MASK))
|
2005-04-16 16:20:36 -06:00
|
|
|
return -1;
|
|
|
|
onh++;
|
|
|
|
} endfor_nexthops(fi);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-01-12 22:49:01 -07:00
|
|
|
static inline unsigned int fib_devindex_hashfn(unsigned int val)
|
|
|
|
{
|
|
|
|
unsigned int mask = DEVINDEX_HASHSIZE - 1;
|
|
|
|
|
|
|
|
return (val ^
|
|
|
|
(val >> DEVINDEX_HASHBITS) ^
|
|
|
|
(val >> (DEVINDEX_HASHBITS * 2))) & mask;
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
static inline unsigned int fib_info_hashfn(const struct fib_info *fi)
|
|
|
|
{
|
2011-02-01 16:34:21 -07:00
|
|
|
unsigned int mask = (fib_info_hash_size - 1);
|
2005-04-16 16:20:36 -06:00
|
|
|
unsigned int val = fi->fib_nhs;
|
|
|
|
|
2011-03-24 19:06:47 -06:00
|
|
|
val ^= (fi->fib_protocol << 8) | fi->fib_scope;
|
2006-09-27 19:40:00 -06:00
|
|
|
val ^= (__force u32)fi->fib_prefsrc;
|
2005-04-16 16:20:36 -06:00
|
|
|
val ^= fi->fib_priority;
|
2008-01-12 22:49:01 -07:00
|
|
|
for_nexthops(fi) {
|
|
|
|
val ^= fib_devindex_hashfn(nh->nh_oif);
|
|
|
|
} endfor_nexthops(fi)
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
return (val ^ (val >> 7) ^ (val >> 12)) & mask;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct fib_info *fib_find_info(const struct fib_info *nfi)
|
|
|
|
{
|
|
|
|
struct hlist_head *head;
|
|
|
|
struct fib_info *fi;
|
|
|
|
unsigned int hash;
|
|
|
|
|
|
|
|
hash = fib_info_hashfn(nfi);
|
|
|
|
head = &fib_info_hash[hash];
|
|
|
|
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-27 18:06:00 -07:00
|
|
|
hlist_for_each_entry(fi, head, fib_hash) {
|
2009-11-25 16:14:13 -07:00
|
|
|
if (!net_eq(fi->fib_net, nfi->fib_net))
|
2008-01-31 19:50:07 -07:00
|
|
|
continue;
|
2005-04-16 16:20:36 -06:00
|
|
|
if (fi->fib_nhs != nfi->fib_nhs)
|
|
|
|
continue;
|
|
|
|
if (nfi->fib_protocol == fi->fib_protocol &&
|
2011-03-24 19:06:47 -06:00
|
|
|
nfi->fib_scope == fi->fib_scope &&
|
2005-04-16 16:20:36 -06:00
|
|
|
nfi->fib_prefsrc == fi->fib_prefsrc &&
|
|
|
|
nfi->fib_priority == fi->fib_priority &&
|
2012-10-03 19:25:26 -06:00
|
|
|
nfi->fib_type == fi->fib_type &&
|
2005-04-16 16:20:36 -06:00
|
|
|
memcmp(nfi->fib_metrics, fi->fib_metrics,
|
2011-03-24 01:01:24 -06:00
|
|
|
sizeof(u32) * RTAX_MAX) == 0 &&
|
2015-06-23 11:45:36 -06:00
|
|
|
!((nfi->fib_flags ^ fi->fib_flags) & ~RTNH_COMPARE_MASK) &&
|
2005-04-16 16:20:36 -06:00
|
|
|
(nfi->fib_nhs == 0 || nh_comp(fi, nfi) == 0))
|
|
|
|
return fi;
|
|
|
|
}
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Check, that the gateway is already configured.
|
2010-10-04 14:00:18 -06:00
|
|
|
* Used only by redirect accept routine.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2006-09-26 23:18:13 -06:00
|
|
|
int ip_fib_check_default(__be32 gw, struct net_device *dev)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct hlist_head *head;
|
|
|
|
struct fib_nh *nh;
|
|
|
|
unsigned int hash;
|
|
|
|
|
2006-08-29 17:48:09 -06:00
|
|
|
spin_lock(&fib_info_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
hash = fib_devindex_hashfn(dev->ifindex);
|
|
|
|
head = &fib_info_devhash[hash];
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-27 18:06:00 -07:00
|
|
|
hlist_for_each_entry(nh, head, nh_hash) {
|
2005-04-16 16:20:36 -06:00
|
|
|
if (nh->nh_dev == dev &&
|
|
|
|
nh->nh_gw == gw &&
|
2010-10-04 14:00:18 -06:00
|
|
|
!(nh->nh_flags & RTNH_F_DEAD)) {
|
2006-08-29 17:48:09 -06:00
|
|
|
spin_unlock(&fib_info_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2006-08-29 17:48:09 -06:00
|
|
|
spin_unlock(&fib_info_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2006-11-10 15:10:15 -07:00
|
|
|
static inline size_t fib_nlmsg_size(struct fib_info *fi)
|
|
|
|
{
|
|
|
|
size_t payload = NLMSG_ALIGN(sizeof(struct rtmsg))
|
|
|
|
+ nla_total_size(4) /* RTA_TABLE */
|
|
|
|
+ nla_total_size(4) /* RTA_DST */
|
|
|
|
+ nla_total_size(4) /* RTA_PRIORITY */
|
2015-01-05 15:57:47 -07:00
|
|
|
+ nla_total_size(4) /* RTA_PREFSRC */
|
|
|
|
+ nla_total_size(TCP_CA_NAME_MAX); /* RTAX_CC_ALGO */
|
2006-11-10 15:10:15 -07:00
|
|
|
|
|
|
|
/* space for nested metrics */
|
|
|
|
payload += nla_total_size((RTAX_MAX * nla_total_size(4)));
|
|
|
|
|
|
|
|
if (fi->fib_nhs) {
|
2015-07-21 02:43:47 -06:00
|
|
|
size_t nh_encapsize = 0;
|
2006-11-10 15:10:15 -07:00
|
|
|
/* Also handles the special case fib_nhs == 1 */
|
|
|
|
|
|
|
|
/* each nexthop is packed in an attribute */
|
|
|
|
size_t nhsize = nla_total_size(sizeof(struct rtnexthop));
|
|
|
|
|
|
|
|
/* may contain flow and gateway attribute */
|
|
|
|
nhsize += 2 * nla_total_size(4);
|
|
|
|
|
2015-07-21 02:43:47 -06:00
|
|
|
/* grab encap info */
|
|
|
|
for_nexthops(fi) {
|
|
|
|
if (nh->nh_lwtstate) {
|
|
|
|
/* RTA_ENCAP_TYPE */
|
|
|
|
nh_encapsize += lwtunnel_get_encap_size(
|
|
|
|
nh->nh_lwtstate);
|
|
|
|
/* RTA_ENCAP */
|
|
|
|
nh_encapsize += nla_total_size(2);
|
|
|
|
}
|
|
|
|
} endfor_nexthops(fi);
|
|
|
|
|
2006-11-10 15:10:15 -07:00
|
|
|
/* all nexthops are packed in a nested attribute */
|
2015-07-21 02:43:47 -06:00
|
|
|
payload += nla_total_size((fi->fib_nhs * nhsize) +
|
|
|
|
nh_encapsize);
|
|
|
|
|
2006-11-10 15:10:15 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
return payload;
|
|
|
|
}
|
|
|
|
|
2006-09-27 19:40:00 -06:00
|
|
|
void rtmsg_fib(int event, __be32 key, struct fib_alias *fa,
|
2013-10-17 14:34:11 -06:00
|
|
|
int dst_len, u32 tb_id, const struct nl_info *info,
|
2007-05-23 15:55:06 -06:00
|
|
|
unsigned int nlm_flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct sk_buff *skb;
|
2006-08-17 19:14:52 -06:00
|
|
|
u32 seq = info->nlh ? info->nlh->nlmsg_seq : 0;
|
2006-08-15 01:34:17 -06:00
|
|
|
int err = -ENOBUFS;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-11-10 15:10:15 -07:00
|
|
|
skb = nlmsg_new(fib_nlmsg_size(fa->fa_info), GFP_KERNEL);
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!skb)
|
2006-08-15 01:34:17 -06:00
|
|
|
goto errout;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2012-09-07 14:12:54 -06:00
|
|
|
err = fib_dump_info(skb, info->portid, seq, event, tb_id,
|
2011-03-24 19:06:47 -06:00
|
|
|
fa->fa_type, key, dst_len,
|
2007-05-23 15:55:06 -06:00
|
|
|
fa->fa_tos, fa->fa_info, nlm_flags);
|
2007-02-01 00:16:40 -07:00
|
|
|
if (err < 0) {
|
|
|
|
/* -EMSGSIZE implies BUG in fib_nlmsg_size() */
|
|
|
|
WARN_ON(err == -EMSGSIZE);
|
|
|
|
kfree_skb(skb);
|
|
|
|
goto errout;
|
|
|
|
}
|
2012-09-07 14:12:54 -06:00
|
|
|
rtnl_notify(skb, info->nl_net, info->portid, RTNLGRP_IPV4_ROUTE,
|
2009-02-25 00:18:28 -07:00
|
|
|
info->nlh, GFP_KERNEL);
|
|
|
|
return;
|
2006-08-15 01:34:17 -06:00
|
|
|
errout:
|
|
|
|
if (err < 0)
|
2008-01-10 04:26:13 -07:00
|
|
|
rtnl_set_sk_err(info->nl_net, RTNLGRP_IPV4_ROUTE, err);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2013-12-28 12:05:36 -07:00
|
|
|
static int fib_detect_death(struct fib_info *fi, int order,
|
|
|
|
struct fib_info **last_resort, int *last_idx,
|
|
|
|
int dflt)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct neighbour *n;
|
|
|
|
int state = NUD_NONE;
|
|
|
|
|
|
|
|
n = neigh_lookup(&arp_tbl, &fi->fib_nh[0].nh_gw, fi->fib_dev);
|
|
|
|
if (n) {
|
|
|
|
state = n->nud_state;
|
|
|
|
neigh_release(n);
|
2015-07-23 01:39:35 -06:00
|
|
|
} else {
|
|
|
|
return 0;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2008-11-03 01:23:42 -07:00
|
|
|
if (state == NUD_REACHABLE)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
2010-10-04 14:00:18 -06:00
|
|
|
if ((state & NUD_VALID) && order != dflt)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
2010-10-04 14:00:18 -06:00
|
|
|
if ((state & NUD_VALID) ||
|
2015-07-23 01:39:35 -06:00
|
|
|
(*last_idx < 0 && order > dflt && state != NUD_INCOMPLETE)) {
|
2005-04-16 16:20:36 -06:00
|
|
|
*last_resort = fi;
|
|
|
|
*last_idx = order;
|
|
|
|
}
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
static int fib_count_nexthops(struct rtnexthop *rtnh, int remaining)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
int nhs = 0;
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
while (rtnh_ok(rtnh, remaining)) {
|
2005-04-16 16:20:36 -06:00
|
|
|
nhs++;
|
2006-08-17 19:14:52 -06:00
|
|
|
rtnh = rtnh_next(rtnh, &remaining);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* leftover implies invalid nexthop configuration, discard it */
|
|
|
|
return remaining > 0 ? 0 : nhs;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
|
|
|
|
int remaining, struct fib_config *cfg)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2015-07-21 02:43:47 -06:00
|
|
|
int ret;
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
change_nexthops(fi) {
|
2006-08-17 19:14:52 -06:00
|
|
|
int attrlen;
|
|
|
|
|
|
|
|
if (!rtnh_ok(rtnh, remaining))
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EINVAL;
|
2006-08-17 19:14:52 -06:00
|
|
|
|
2016-07-10 12:11:55 -06:00
|
|
|
if (rtnh->rtnh_flags & (RTNH_F_DEAD | RTNH_F_LINKDOWN))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2010-01-15 02:16:40 -07:00
|
|
|
nexthop_nh->nh_flags =
|
|
|
|
(cfg->fc_flags & ~0xFF) | rtnh->rtnh_flags;
|
|
|
|
nexthop_nh->nh_oif = rtnh->rtnh_ifindex;
|
|
|
|
nexthop_nh->nh_weight = rtnh->rtnh_hops + 1;
|
2006-08-17 19:14:52 -06:00
|
|
|
|
|
|
|
attrlen = rtnh_attrlen(rtnh);
|
|
|
|
if (attrlen > 0) {
|
|
|
|
struct nlattr *nla, *attrs = rtnh_attrs(rtnh);
|
|
|
|
|
|
|
|
nla = nla_find(attrs, attrlen, RTA_GATEWAY);
|
2015-03-29 08:59:26 -06:00
|
|
|
nexthop_nh->nh_gw = nla ? nla_get_in_addr(nla) : 0;
|
2011-01-14 05:36:42 -07:00
|
|
|
#ifdef CONFIG_IP_ROUTE_CLASSID
|
2006-08-17 19:14:52 -06:00
|
|
|
nla = nla_find(attrs, attrlen, RTA_FLOW);
|
2010-01-15 02:16:40 -07:00
|
|
|
nexthop_nh->nh_tclassid = nla ? nla_get_u32(nla) : 0;
|
2012-06-29 02:32:45 -06:00
|
|
|
if (nexthop_nh->nh_tclassid)
|
2012-07-05 23:13:13 -06:00
|
|
|
fi->fib_net->ipv4.fib_num_tclassid_users++;
|
2005-04-16 16:20:36 -06:00
|
|
|
#endif
|
2015-07-21 02:43:47 -06:00
|
|
|
nla = nla_find(attrs, attrlen, RTA_ENCAP);
|
|
|
|
if (nla) {
|
|
|
|
struct lwtunnel_state *lwtstate;
|
|
|
|
struct nlattr *nla_entype;
|
|
|
|
|
|
|
|
nla_entype = nla_find(attrs, attrlen,
|
|
|
|
RTA_ENCAP_TYPE);
|
|
|
|
if (!nla_entype)
|
|
|
|
goto err_inval;
|
2017-01-30 13:07:37 -07:00
|
|
|
|
|
|
|
ret = lwtunnel_build_state(nla_get_u16(
|
2015-07-21 02:43:47 -06:00
|
|
|
nla_entype),
|
2015-08-24 10:45:41 -06:00
|
|
|
nla, AF_INET, cfg,
|
|
|
|
&lwtstate);
|
2015-07-21 02:43:47 -06:00
|
|
|
if (ret)
|
|
|
|
goto errout;
|
2015-07-24 04:28:36 -06:00
|
|
|
nexthop_nh->nh_lwtstate =
|
|
|
|
lwtstate_get(lwtstate);
|
2015-07-21 02:43:47 -06:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2006-08-17 19:14:52 -06:00
|
|
|
|
|
|
|
rtnh = rtnh_next(rtnh, &remaining);
|
2005-04-16 16:20:36 -06:00
|
|
|
} endfor_nexthops(fi);
|
2006-08-17 19:14:52 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
2015-07-21 02:43:47 -06:00
|
|
|
|
|
|
|
err_inval:
|
|
|
|
ret = -EINVAL;
|
|
|
|
|
|
|
|
errout:
|
|
|
|
return ret;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2015-09-30 02:12:21 -06:00
|
|
|
static void fib_rebalance(struct fib_info *fi)
|
|
|
|
{
|
|
|
|
int total;
|
|
|
|
int w;
|
|
|
|
struct in_device *in_dev;
|
|
|
|
|
|
|
|
if (fi->fib_nhs < 2)
|
|
|
|
return;
|
|
|
|
|
|
|
|
total = 0;
|
|
|
|
for_nexthops(fi) {
|
|
|
|
if (nh->nh_flags & RTNH_F_DEAD)
|
|
|
|
continue;
|
|
|
|
|
2015-10-14 17:44:00 -06:00
|
|
|
in_dev = __in_dev_get_rtnl(nh->nh_dev);
|
2015-09-30 02:12:21 -06:00
|
|
|
|
|
|
|
if (in_dev &&
|
|
|
|
IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
|
|
|
|
nh->nh_flags & RTNH_F_LINKDOWN)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
total += nh->nh_weight;
|
|
|
|
} endfor_nexthops(fi);
|
|
|
|
|
|
|
|
w = 0;
|
|
|
|
change_nexthops(fi) {
|
|
|
|
int upper_bound;
|
|
|
|
|
2015-10-14 17:44:00 -06:00
|
|
|
in_dev = __in_dev_get_rtnl(nexthop_nh->nh_dev);
|
2015-09-30 02:12:21 -06:00
|
|
|
|
|
|
|
if (nexthop_nh->nh_flags & RTNH_F_DEAD) {
|
|
|
|
upper_bound = -1;
|
|
|
|
} else if (in_dev &&
|
|
|
|
IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
|
|
|
|
nexthop_nh->nh_flags & RTNH_F_LINKDOWN) {
|
|
|
|
upper_bound = -1;
|
|
|
|
} else {
|
|
|
|
w += nexthop_nh->nh_weight;
|
2015-10-05 23:24:47 -06:00
|
|
|
upper_bound = DIV_ROUND_CLOSEST_ULL((u64)w << 31,
|
|
|
|
total) - 1;
|
2015-09-30 02:12:21 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
atomic_set(&nexthop_nh->nh_upper_bound, upper_bound);
|
|
|
|
} endfor_nexthops(fi);
|
|
|
|
|
|
|
|
net_get_random_once(&fib_multipath_secret,
|
|
|
|
sizeof(fib_multipath_secret));
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void fib_add_weight(struct fib_info *fi,
|
|
|
|
const struct fib_nh *nh)
|
|
|
|
{
|
|
|
|
fi->fib_weight += nh->nh_weight;
|
|
|
|
}
|
|
|
|
|
|
|
|
#else /* CONFIG_IP_ROUTE_MULTIPATH */
|
|
|
|
|
|
|
|
#define fib_rebalance(fi) do { } while (0)
|
|
|
|
#define fib_add_weight(fi, nh) do { } while (0)
|
|
|
|
|
|
|
|
#endif /* CONFIG_IP_ROUTE_MULTIPATH */
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2017-01-30 13:07:37 -07:00
|
|
|
static int fib_encap_match(u16 encap_type,
|
2015-08-19 02:04:51 -06:00
|
|
|
struct nlattr *encap,
|
2017-01-30 13:07:37 -07:00
|
|
|
const struct fib_nh *nh,
|
2015-08-24 10:45:41 -06:00
|
|
|
const struct fib_config *cfg)
|
2015-07-21 02:43:47 -06:00
|
|
|
{
|
|
|
|
struct lwtunnel_state *lwtstate;
|
2015-08-18 10:41:13 -06:00
|
|
|
int ret, result = 0;
|
2015-07-21 02:43:47 -06:00
|
|
|
|
|
|
|
if (encap_type == LWTUNNEL_ENCAP_NONE)
|
|
|
|
return 0;
|
|
|
|
|
2017-01-30 13:07:37 -07:00
|
|
|
ret = lwtunnel_build_state(encap_type, encap,
|
2015-08-24 10:45:41 -06:00
|
|
|
AF_INET, cfg, &lwtstate);
|
2015-08-18 10:41:13 -06:00
|
|
|
if (!ret) {
|
|
|
|
result = lwtunnel_cmp_encap(lwtstate, nh->nh_lwtstate);
|
|
|
|
lwtstate_free(lwtstate);
|
|
|
|
}
|
2015-07-21 02:43:47 -06:00
|
|
|
|
2015-08-18 10:41:13 -06:00
|
|
|
return result;
|
2015-07-21 02:43:47 -06:00
|
|
|
}
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
int fib_nh_match(struct fib_config *cfg, struct fib_info *fi)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
2006-08-17 19:14:52 -06:00
|
|
|
struct rtnexthop *rtnh;
|
|
|
|
int remaining;
|
2005-04-16 16:20:36 -06:00
|
|
|
#endif
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_priority && cfg->fc_priority != fi->fib_priority)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 1;
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_oif || cfg->fc_gw) {
|
2015-07-21 02:43:47 -06:00
|
|
|
if (cfg->fc_encap) {
|
2017-01-30 13:07:37 -07:00
|
|
|
if (fib_encap_match(cfg->fc_encap_type,
|
|
|
|
cfg->fc_encap, fi->fib_nh, cfg))
|
2015-07-21 02:43:47 -06:00
|
|
|
return 1;
|
|
|
|
}
|
2006-08-17 19:14:52 -06:00
|
|
|
if ((!cfg->fc_oif || cfg->fc_oif == fi->fib_nh->nh_oif) &&
|
|
|
|
(!cfg->fc_gw || cfg->fc_gw == fi->fib_nh->nh_gw))
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!cfg->fc_mp)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
2006-08-17 19:14:52 -06:00
|
|
|
|
|
|
|
rtnh = cfg->fc_mp;
|
|
|
|
remaining = cfg->fc_mp_len;
|
2007-02-09 07:24:47 -07:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
for_nexthops(fi) {
|
2006-08-17 19:14:52 -06:00
|
|
|
int attrlen;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
if (!rtnh_ok(rtnh, remaining))
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EINVAL;
|
2006-08-17 19:14:52 -06:00
|
|
|
|
|
|
|
if (rtnh->rtnh_ifindex && rtnh->rtnh_ifindex != nh->nh_oif)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 1;
|
2006-08-17 19:14:52 -06:00
|
|
|
|
|
|
|
attrlen = rtnh_attrlen(rtnh);
|
2014-10-13 08:34:10 -06:00
|
|
|
if (attrlen > 0) {
|
2006-08-17 19:14:52 -06:00
|
|
|
struct nlattr *nla, *attrs = rtnh_attrs(rtnh);
|
|
|
|
|
|
|
|
nla = nla_find(attrs, attrlen, RTA_GATEWAY);
|
2015-03-29 08:59:26 -06:00
|
|
|
if (nla && nla_get_in_addr(nla) != nh->nh_gw)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 1;
|
2011-01-14 05:36:42 -07:00
|
|
|
#ifdef CONFIG_IP_ROUTE_CLASSID
|
2006-08-17 19:14:52 -06:00
|
|
|
nla = nla_find(attrs, attrlen, RTA_FLOW);
|
|
|
|
if (nla && nla_get_u32(nla) != nh->nh_tclassid)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 1;
|
|
|
|
#endif
|
|
|
|
}
|
2006-08-17 19:14:52 -06:00
|
|
|
|
|
|
|
rtnh = rtnh_next(rtnh, &remaining);
|
2005-04-16 16:20:36 -06:00
|
|
|
} endfor_nexthops(fi);
|
|
|
|
#endif
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/*
|
2010-10-04 14:00:18 -06:00
|
|
|
* Picture
|
|
|
|
* -------
|
|
|
|
*
|
|
|
|
* Semantics of nexthop is very messy by historical reasons.
|
|
|
|
* We have to take into account, that:
|
|
|
|
* a) gateway can be actually local interface address,
|
|
|
|
* so that gatewayed route is direct.
|
|
|
|
* b) gateway must be on-link address, possibly
|
|
|
|
* described not by an ifaddr, but also by a direct route.
|
|
|
|
* c) If both gateway and interface are specified, they should not
|
|
|
|
* contradict.
|
|
|
|
* d) If we use tunnel routes, gateway could be not on-link.
|
|
|
|
*
|
|
|
|
* Attempt to reconcile all of these (alas, self-contradictory) conditions
|
|
|
|
* results in pretty ugly and hairy code with obscure logic.
|
|
|
|
*
|
|
|
|
* I chose to generalized it instead, so that the size
|
|
|
|
* of code does not increase practically, but it becomes
|
|
|
|
* much more general.
|
|
|
|
* Every prefix is assigned a "scope" value: "host" is local address,
|
|
|
|
* "link" is direct route,
|
|
|
|
* [ ... "site" ... "interior" ... ]
|
|
|
|
* and "universe" is true gateway route with global meaning.
|
|
|
|
*
|
|
|
|
* Every prefix refers to a set of "nexthop"s (gw, oif),
|
|
|
|
* where gw must have narrower scope. This recursion stops
|
|
|
|
* when gw has LOCAL scope or if "nexthop" is declared ONLINK,
|
|
|
|
* which means that gw is forced to be on link.
|
|
|
|
*
|
|
|
|
* Code is still hairy, but now it is apparently logically
|
|
|
|
* consistent and very flexible. F.e. as by-product it allows
|
|
|
|
* to co-exists in peace independent exterior and interior
|
|
|
|
* routing processes.
|
|
|
|
*
|
|
|
|
* Normally it looks as following.
|
|
|
|
*
|
|
|
|
* {universe prefix} -> (gw, oif) [scope link]
|
|
|
|
* |
|
|
|
|
* |-> {link prefix} -> (gw, oif) [scope local]
|
|
|
|
* |
|
|
|
|
* |-> {local prefix} (terminal node)
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2006-08-17 19:14:52 -06:00
|
|
|
static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi,
|
|
|
|
struct fib_nh *nh)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2015-08-24 10:45:41 -06:00
|
|
|
int err = 0;
|
2008-01-21 18:34:00 -07:00
|
|
|
struct net *net;
|
2010-10-04 14:00:18 -06:00
|
|
|
struct net_device *dev;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-01-21 18:34:00 -07:00
|
|
|
net = cfg->fc_nlinfo.nl_net;
|
2005-04-16 16:20:36 -06:00
|
|
|
if (nh->nh_gw) {
|
|
|
|
struct fib_result res;
|
|
|
|
|
2010-10-04 14:00:18 -06:00
|
|
|
if (nh->nh_flags & RTNH_F_ONLINK) {
|
2015-08-13 14:59:05 -06:00
|
|
|
unsigned int addr_type;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_scope >= RT_SCOPE_LINK)
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EINVAL;
|
2010-10-04 14:00:18 -06:00
|
|
|
dev = __dev_get_by_index(net, nh->nh_oif);
|
|
|
|
if (!dev)
|
2005-04-16 16:20:36 -06:00
|
|
|
return -ENODEV;
|
2010-10-04 14:00:18 -06:00
|
|
|
if (!(dev->flags & IFF_UP))
|
2005-04-16 16:20:36 -06:00
|
|
|
return -ENETDOWN;
|
2015-08-13 14:59:05 -06:00
|
|
|
addr_type = inet_addr_type_dev_table(net, dev, nh->nh_gw);
|
|
|
|
if (addr_type != RTN_UNICAST)
|
|
|
|
return -EINVAL;
|
2015-06-23 11:45:36 -06:00
|
|
|
if (!netif_carrier_ok(dev))
|
|
|
|
nh->nh_flags |= RTNH_F_LINKDOWN;
|
2005-04-16 16:20:36 -06:00
|
|
|
nh->nh_dev = dev;
|
|
|
|
dev_hold(dev);
|
|
|
|
nh->nh_scope = RT_SCOPE_LINK;
|
|
|
|
return 0;
|
|
|
|
}
|
2010-10-05 04:41:36 -06:00
|
|
|
rcu_read_lock();
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2015-08-13 14:59:07 -06:00
|
|
|
struct fib_table *tbl = NULL;
|
2011-03-12 00:02:42 -07:00
|
|
|
struct flowi4 fl4 = {
|
|
|
|
.daddr = nh->nh_gw,
|
|
|
|
.flowi4_scope = cfg->fc_scope + 1,
|
|
|
|
.flowi4_oif = nh->nh_oif,
|
ipv4, fib: pass LOOPBACK_IFINDEX instead of 0 to flowi4_iif
As suggested by Julian:
Simply, flowi4_iif must not contain 0, it does not
look logical to ignore all ip rules with specified iif.
because in fib_rule_match() we do:
if (rule->iifindex && (rule->iifindex != fl->flowi_iif))
goto out;
flowi4_iif should be LOOPBACK_IFINDEX by default.
We need to move LOOPBACK_IFINDEX to include/net/flow.h:
1) It is mostly used by flowi_iif
2) Fix the following compile error if we use it in flow.h
by the patches latter:
In file included from include/linux/netfilter.h:277:0,
from include/net/netns/netfilter.h:5,
from include/net/net_namespace.h:21,
from include/linux/netdevice.h:43,
from include/linux/icmpv6.h:12,
from include/linux/ipv6.h:61,
from include/net/ipv6.h:16,
from include/linux/sunrpc/clnt.h:27,
from include/linux/nfs_fs.h:30,
from init/do_mounts.c:32:
include/net/flow.h: In function ‘flowi4_init_output’:
include/net/flow.h:84:32: error: ‘LOOPBACK_IFINDEX’ undeclared (first use in this function)
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Julian Anastasov <ja@ssi.bg>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Cong Wang <cwang@twopensource.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-15 17:25:34 -06:00
|
|
|
.flowi4_iif = LOOPBACK_IFINDEX,
|
2006-08-17 19:14:52 -06:00
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* It is not necessary, but requires a bit of thinking */
|
2011-03-12 00:02:42 -07:00
|
|
|
if (fl4.flowi4_scope < RT_SCOPE_LINK)
|
|
|
|
fl4.flowi4_scope = RT_SCOPE_LINK;
|
2015-08-13 14:59:07 -06:00
|
|
|
|
|
|
|
if (cfg->fc_table)
|
|
|
|
tbl = fib_get_table(net, cfg->fc_table);
|
|
|
|
|
|
|
|
if (tbl)
|
|
|
|
err = fib_table_lookup(tbl, &fl4, &res,
|
2015-08-15 11:54:07 -06:00
|
|
|
FIB_LOOKUP_IGNORE_LINKSTATE |
|
|
|
|
FIB_LOOKUP_NOREF);
|
2015-08-19 12:40:31 -06:00
|
|
|
|
|
|
|
/* on error or if no table given do full lookup. This
|
|
|
|
* is needed for example when nexthops are in the local
|
|
|
|
* table rather than the given table
|
|
|
|
*/
|
|
|
|
if (!tbl || err) {
|
2015-08-13 14:59:07 -06:00
|
|
|
err = fib_lookup(net, &fl4, &res,
|
|
|
|
FIB_LOOKUP_IGNORE_LINKSTATE);
|
2015-08-19 12:40:31 -06:00
|
|
|
}
|
|
|
|
|
2010-10-05 04:41:36 -06:00
|
|
|
if (err) {
|
|
|
|
rcu_read_unlock();
|
2005-04-16 16:20:36 -06:00
|
|
|
return err;
|
2010-10-05 04:41:36 -06:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
err = -EINVAL;
|
|
|
|
if (res.type != RTN_UNICAST && res.type != RTN_LOCAL)
|
|
|
|
goto out;
|
|
|
|
nh->nh_scope = res.scope;
|
|
|
|
nh->nh_oif = FIB_RES_OIF(res);
|
2010-10-04 14:00:18 -06:00
|
|
|
nh->nh_dev = dev = FIB_RES_DEV(res);
|
|
|
|
if (!dev)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
2010-10-04 14:00:18 -06:00
|
|
|
dev_hold(dev);
|
2015-06-23 11:45:36 -06:00
|
|
|
if (!netif_carrier_ok(dev))
|
|
|
|
nh->nh_flags |= RTNH_F_LINKDOWN;
|
2010-10-18 18:39:26 -06:00
|
|
|
err = (dev->flags & IFF_UP) ? 0 : -ENETDOWN;
|
2005-04-16 16:20:36 -06:00
|
|
|
} else {
|
|
|
|
struct in_device *in_dev;
|
|
|
|
|
2010-10-04 14:00:18 -06:00
|
|
|
if (nh->nh_flags & (RTNH_F_PERVASIVE | RTNH_F_ONLINK))
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EINVAL;
|
|
|
|
|
2010-10-18 18:39:26 -06:00
|
|
|
rcu_read_lock();
|
|
|
|
err = -ENODEV;
|
2008-01-21 18:34:00 -07:00
|
|
|
in_dev = inetdev_by_index(net, nh->nh_oif);
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!in_dev)
|
2010-10-18 18:39:26 -06:00
|
|
|
goto out;
|
|
|
|
err = -ENETDOWN;
|
|
|
|
if (!(in_dev->dev->flags & IFF_UP))
|
|
|
|
goto out;
|
2005-04-16 16:20:36 -06:00
|
|
|
nh->nh_dev = in_dev->dev;
|
|
|
|
dev_hold(nh->nh_dev);
|
|
|
|
nh->nh_scope = RT_SCOPE_HOST;
|
2015-06-23 11:45:36 -06:00
|
|
|
if (!netif_carrier_ok(nh->nh_dev))
|
|
|
|
nh->nh_flags |= RTNH_F_LINKDOWN;
|
2010-10-18 18:39:26 -06:00
|
|
|
err = 0;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2010-10-18 18:39:26 -06:00
|
|
|
out:
|
|
|
|
rcu_read_unlock();
|
|
|
|
return err;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-09-27 19:40:00 -06:00
|
|
|
static inline unsigned int fib_laddr_hashfn(__be32 val)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2011-02-01 16:34:21 -07:00
|
|
|
unsigned int mask = (fib_info_hash_size - 1);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-10-04 14:00:18 -06:00
|
|
|
return ((__force u32)val ^
|
|
|
|
((__force u32)val >> 7) ^
|
|
|
|
((__force u32)val >> 14)) & mask;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2011-02-01 16:34:21 -07:00
|
|
|
static struct hlist_head *fib_info_hash_alloc(int bytes)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
if (bytes <= PAGE_SIZE)
|
2007-11-26 08:29:32 -07:00
|
|
|
return kzalloc(bytes, GFP_KERNEL);
|
2005-04-16 16:20:36 -06:00
|
|
|
else
|
|
|
|
return (struct hlist_head *)
|
2010-10-04 14:00:18 -06:00
|
|
|
__get_free_pages(GFP_KERNEL | __GFP_ZERO,
|
|
|
|
get_order(bytes));
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2011-02-01 16:34:21 -07:00
|
|
|
static void fib_info_hash_free(struct hlist_head *hash, int bytes)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
if (!hash)
|
|
|
|
return;
|
|
|
|
|
|
|
|
if (bytes <= PAGE_SIZE)
|
|
|
|
kfree(hash);
|
|
|
|
else
|
|
|
|
free_pages((unsigned long) hash, get_order(bytes));
|
|
|
|
}
|
|
|
|
|
2011-02-01 16:34:21 -07:00
|
|
|
static void fib_info_hash_move(struct hlist_head *new_info_hash,
|
|
|
|
struct hlist_head *new_laddrhash,
|
|
|
|
unsigned int new_size)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2005-08-05 05:12:48 -06:00
|
|
|
struct hlist_head *old_info_hash, *old_laddrhash;
|
2011-02-01 16:34:21 -07:00
|
|
|
unsigned int old_size = fib_info_hash_size;
|
2005-08-05 05:12:48 -06:00
|
|
|
unsigned int i, bytes;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-29 17:48:09 -06:00
|
|
|
spin_lock_bh(&fib_info_lock);
|
2005-08-05 05:12:48 -06:00
|
|
|
old_info_hash = fib_info_hash;
|
|
|
|
old_laddrhash = fib_info_laddrhash;
|
2011-02-01 16:34:21 -07:00
|
|
|
fib_info_hash_size = new_size;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
for (i = 0; i < old_size; i++) {
|
|
|
|
struct hlist_head *head = &fib_info_hash[i];
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-27 18:06:00 -07:00
|
|
|
struct hlist_node *n;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct fib_info *fi;
|
|
|
|
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-27 18:06:00 -07:00
|
|
|
hlist_for_each_entry_safe(fi, n, head, fib_hash) {
|
2005-04-16 16:20:36 -06:00
|
|
|
struct hlist_head *dest;
|
|
|
|
unsigned int new_hash;
|
|
|
|
|
|
|
|
new_hash = fib_info_hashfn(fi);
|
|
|
|
dest = &new_info_hash[new_hash];
|
|
|
|
hlist_add_head(&fi->fib_hash, dest);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
fib_info_hash = new_info_hash;
|
|
|
|
|
|
|
|
for (i = 0; i < old_size; i++) {
|
|
|
|
struct hlist_head *lhead = &fib_info_laddrhash[i];
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-27 18:06:00 -07:00
|
|
|
struct hlist_node *n;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct fib_info *fi;
|
|
|
|
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-27 18:06:00 -07:00
|
|
|
hlist_for_each_entry_safe(fi, n, lhead, fib_lhash) {
|
2005-04-16 16:20:36 -06:00
|
|
|
struct hlist_head *ldest;
|
|
|
|
unsigned int new_hash;
|
|
|
|
|
|
|
|
new_hash = fib_laddr_hashfn(fi->fib_prefsrc);
|
|
|
|
ldest = &new_laddrhash[new_hash];
|
|
|
|
hlist_add_head(&fi->fib_lhash, ldest);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
fib_info_laddrhash = new_laddrhash;
|
|
|
|
|
2006-08-29 17:48:09 -06:00
|
|
|
spin_unlock_bh(&fib_info_lock);
|
2005-08-05 05:12:48 -06:00
|
|
|
|
|
|
|
bytes = old_size * sizeof(struct hlist_head *);
|
2011-02-01 16:34:21 -07:00
|
|
|
fib_info_hash_free(old_info_hash, bytes);
|
|
|
|
fib_info_hash_free(old_laddrhash, bytes);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2011-03-24 18:42:21 -06:00
|
|
|
__be32 fib_info_update_nh_saddr(struct net *net, struct fib_nh *nh)
|
|
|
|
{
|
|
|
|
nh->nh_saddr = inet_select_addr(nh->nh_dev,
|
|
|
|
nh->nh_gw,
|
2011-03-24 19:06:47 -06:00
|
|
|
nh->nh_parent->fib_scope);
|
2011-03-24 18:42:21 -06:00
|
|
|
nh->nh_saddr_genid = atomic_read(&net->ipv4.dev_addr_genid);
|
|
|
|
|
|
|
|
return nh->nh_saddr;
|
|
|
|
}
|
|
|
|
|
2015-08-13 14:59:06 -06:00
|
|
|
static bool fib_valid_prefsrc(struct fib_config *cfg, __be32 fib_prefsrc)
|
|
|
|
{
|
|
|
|
if (cfg->fc_type != RTN_LOCAL || !cfg->fc_dst ||
|
|
|
|
fib_prefsrc != cfg->fc_dst) {
|
2015-09-01 14:26:35 -06:00
|
|
|
u32 tb_id = cfg->fc_table;
|
2015-11-03 16:59:28 -07:00
|
|
|
int rc;
|
2015-08-13 14:59:06 -06:00
|
|
|
|
|
|
|
if (tb_id == RT_TABLE_MAIN)
|
|
|
|
tb_id = RT_TABLE_LOCAL;
|
|
|
|
|
2015-11-03 16:59:28 -07:00
|
|
|
rc = inet_addr_type_table(cfg->fc_nlinfo.nl_net,
|
|
|
|
fib_prefsrc, tb_id);
|
|
|
|
|
|
|
|
if (rc != RTN_LOCAL && tb_id != RT_TABLE_LOCAL) {
|
|
|
|
rc = inet_addr_type_table(cfg->fc_nlinfo.nl_net,
|
|
|
|
fib_prefsrc, RT_TABLE_LOCAL);
|
2015-08-13 14:59:06 -06:00
|
|
|
}
|
2015-11-03 16:59:28 -07:00
|
|
|
|
|
|
|
if (rc != RTN_LOCAL)
|
|
|
|
return false;
|
2015-08-13 14:59:06 -06:00
|
|
|
}
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2015-08-31 07:58:44 -06:00
|
|
|
static int
|
|
|
|
fib_convert_metrics(struct fib_info *fi, const struct fib_config *cfg)
|
|
|
|
{
|
tcp: use dctcp if enabled on the route to the initiator
Currently, the following case doesn't use DCTCP, even if it should:
A responder has f.e. Cubic as system wide default, but for a specific
route to the initiating host, DCTCP is being set in RTAX_CC_ALGO. The
initiating host then uses DCTCP as congestion control, but since the
initiator sets ECT(0), tcp_ecn_create_request() doesn't set ecn_ok,
and we have to fall back to Reno after 3WHS completes.
We were thinking on how to solve this in a minimal, non-intrusive
way without bloating tcp_ecn_create_request() needlessly: lets cache
the CA ecn option flag in RTAX_FEATURES. In other words, when ECT(0)
is set on the SYN packet, set ecn_ok=1 iff route RTAX_FEATURES
contains the unexposed (internal-only) DST_FEATURE_ECN_CA. This allows
to only do a single metric feature lookup inside tcp_ecn_create_request().
Joint work with Florian Westphal.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-31 07:58:47 -06:00
|
|
|
bool ecn_ca = false;
|
2015-08-31 07:58:44 -06:00
|
|
|
struct nlattr *nla;
|
|
|
|
int remaining;
|
|
|
|
|
|
|
|
if (!cfg->fc_mx)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
nla_for_each_attr(nla, cfg->fc_mx, cfg->fc_mx_len, remaining) {
|
|
|
|
int type = nla_type(nla);
|
|
|
|
u32 val;
|
|
|
|
|
|
|
|
if (!type)
|
|
|
|
continue;
|
|
|
|
if (type > RTAX_MAX)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (type == RTAX_CC_ALGO) {
|
|
|
|
char tmp[TCP_CA_NAME_MAX];
|
|
|
|
|
|
|
|
nla_strlcpy(tmp, nla, sizeof(tmp));
|
tcp: use dctcp if enabled on the route to the initiator
Currently, the following case doesn't use DCTCP, even if it should:
A responder has f.e. Cubic as system wide default, but for a specific
route to the initiating host, DCTCP is being set in RTAX_CC_ALGO. The
initiating host then uses DCTCP as congestion control, but since the
initiator sets ECT(0), tcp_ecn_create_request() doesn't set ecn_ok,
and we have to fall back to Reno after 3WHS completes.
We were thinking on how to solve this in a minimal, non-intrusive
way without bloating tcp_ecn_create_request() needlessly: lets cache
the CA ecn option flag in RTAX_FEATURES. In other words, when ECT(0)
is set on the SYN packet, set ecn_ok=1 iff route RTAX_FEATURES
contains the unexposed (internal-only) DST_FEATURE_ECN_CA. This allows
to only do a single metric feature lookup inside tcp_ecn_create_request().
Joint work with Florian Westphal.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-31 07:58:47 -06:00
|
|
|
val = tcp_ca_get_key_by_name(tmp, &ecn_ca);
|
2015-08-31 07:58:44 -06:00
|
|
|
if (val == TCP_CA_UNSPEC)
|
|
|
|
return -EINVAL;
|
|
|
|
} else {
|
|
|
|
val = nla_get_u32(nla);
|
|
|
|
}
|
|
|
|
if (type == RTAX_ADVMSS && val > 65535 - 40)
|
|
|
|
val = 65535 - 40;
|
|
|
|
if (type == RTAX_MTU && val > 65535 - 15)
|
|
|
|
val = 65535 - 15;
|
2016-05-13 10:33:41 -06:00
|
|
|
if (type == RTAX_HOPLIMIT && val > 255)
|
|
|
|
val = 255;
|
2015-08-31 07:58:46 -06:00
|
|
|
if (type == RTAX_FEATURES && (val & ~RTAX_FEATURE_MASK))
|
|
|
|
return -EINVAL;
|
2015-08-31 07:58:44 -06:00
|
|
|
fi->fib_metrics[type - 1] = val;
|
|
|
|
}
|
|
|
|
|
tcp: use dctcp if enabled on the route to the initiator
Currently, the following case doesn't use DCTCP, even if it should:
A responder has f.e. Cubic as system wide default, but for a specific
route to the initiating host, DCTCP is being set in RTAX_CC_ALGO. The
initiating host then uses DCTCP as congestion control, but since the
initiator sets ECT(0), tcp_ecn_create_request() doesn't set ecn_ok,
and we have to fall back to Reno after 3WHS completes.
We were thinking on how to solve this in a minimal, non-intrusive
way without bloating tcp_ecn_create_request() needlessly: lets cache
the CA ecn option flag in RTAX_FEATURES. In other words, when ECT(0)
is set on the SYN packet, set ecn_ok=1 iff route RTAX_FEATURES
contains the unexposed (internal-only) DST_FEATURE_ECN_CA. This allows
to only do a single metric feature lookup inside tcp_ecn_create_request().
Joint work with Florian Westphal.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-31 07:58:47 -06:00
|
|
|
if (ecn_ca)
|
|
|
|
fi->fib_metrics[RTAX_FEATURES - 1] |= DST_FEATURE_ECN_CA;
|
|
|
|
|
2015-08-31 07:58:44 -06:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
struct fib_info *fib_create_info(struct fib_config *cfg)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
int err;
|
|
|
|
struct fib_info *fi = NULL;
|
|
|
|
struct fib_info *ofi;
|
|
|
|
int nhs = 1;
|
2008-01-31 19:49:32 -07:00
|
|
|
struct net *net = cfg->fc_nlinfo.nl_net;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-03-07 15:27:38 -07:00
|
|
|
if (cfg->fc_type > RTN_MAX)
|
|
|
|
goto err_inval;
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/* Fast check to catch the most weird cases */
|
2006-08-17 19:14:52 -06:00
|
|
|
if (fib_props[cfg->fc_type].scope > cfg->fc_scope)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto err_inval;
|
|
|
|
|
2016-07-10 12:11:55 -06:00
|
|
|
if (cfg->fc_flags & (RTNH_F_DEAD | RTNH_F_LINKDOWN))
|
|
|
|
goto err_inval;
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_mp) {
|
|
|
|
nhs = fib_count_nexthops(cfg->fc_mp, cfg->fc_mp_len);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (nhs == 0)
|
|
|
|
goto err_inval;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
err = -ENOBUFS;
|
2011-02-01 16:34:21 -07:00
|
|
|
if (fib_info_cnt >= fib_info_hash_size) {
|
|
|
|
unsigned int new_size = fib_info_hash_size << 1;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct hlist_head *new_info_hash;
|
|
|
|
struct hlist_head *new_laddrhash;
|
|
|
|
unsigned int bytes;
|
|
|
|
|
|
|
|
if (!new_size)
|
2012-10-21 14:12:09 -06:00
|
|
|
new_size = 16;
|
2005-04-16 16:20:36 -06:00
|
|
|
bytes = new_size * sizeof(struct hlist_head *);
|
2011-02-01 16:34:21 -07:00
|
|
|
new_info_hash = fib_info_hash_alloc(bytes);
|
|
|
|
new_laddrhash = fib_info_hash_alloc(bytes);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (!new_info_hash || !new_laddrhash) {
|
2011-02-01 16:34:21 -07:00
|
|
|
fib_info_hash_free(new_info_hash, bytes);
|
|
|
|
fib_info_hash_free(new_laddrhash, bytes);
|
2007-11-26 08:29:32 -07:00
|
|
|
} else
|
2011-02-01 16:34:21 -07:00
|
|
|
fib_info_hash_move(new_info_hash, new_laddrhash, new_size);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-02-01 16:34:21 -07:00
|
|
|
if (!fib_info_hash_size)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto failure;
|
|
|
|
}
|
|
|
|
|
2006-07-21 15:51:30 -06:00
|
|
|
fi = kzalloc(sizeof(*fi)+nhs*sizeof(struct fib_nh), GFP_KERNEL);
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!fi)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto failure;
|
2014-05-06 09:23:08 -06:00
|
|
|
fib_info_cnt++;
|
2011-01-28 15:05:05 -07:00
|
|
|
if (cfg->fc_mx) {
|
|
|
|
fi->fib_metrics = kzalloc(sizeof(u32) * RTAX_MAX, GFP_KERNEL);
|
|
|
|
if (!fi->fib_metrics)
|
|
|
|
goto failure;
|
|
|
|
} else
|
|
|
|
fi->fib_metrics = (u32 *) dst_default_metrics;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2015-03-11 22:04:08 -06:00
|
|
|
fi->fib_net = net;
|
2006-08-17 19:14:52 -06:00
|
|
|
fi->fib_protocol = cfg->fc_protocol;
|
2011-03-24 19:06:47 -06:00
|
|
|
fi->fib_scope = cfg->fc_scope;
|
2006-08-17 19:14:52 -06:00
|
|
|
fi->fib_flags = cfg->fc_flags;
|
|
|
|
fi->fib_priority = cfg->fc_priority;
|
|
|
|
fi->fib_prefsrc = cfg->fc_prefsrc;
|
2012-10-03 19:25:26 -06:00
|
|
|
fi->fib_type = cfg->fc_type;
|
2016-09-04 16:20:20 -06:00
|
|
|
fi->fib_tb_id = cfg->fc_table;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
fi->fib_nhs = nhs;
|
|
|
|
change_nexthops(fi) {
|
2010-01-15 02:16:40 -07:00
|
|
|
nexthop_nh->nh_parent = fi;
|
2012-07-30 23:45:30 -06:00
|
|
|
nexthop_nh->nh_pcpu_rth_output = alloc_percpu(struct rtable __rcu *);
|
2012-10-08 05:41:17 -06:00
|
|
|
if (!nexthop_nh->nh_pcpu_rth_output)
|
|
|
|
goto failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
} endfor_nexthops(fi)
|
|
|
|
|
2015-08-31 07:58:44 -06:00
|
|
|
err = fib_convert_metrics(fi, cfg);
|
|
|
|
if (err)
|
|
|
|
goto failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_mp) {
|
2005-04-16 16:20:36 -06:00
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
2006-08-17 19:14:52 -06:00
|
|
|
err = fib_get_nhs(fi, cfg->fc_mp, cfg->fc_mp_len, cfg);
|
|
|
|
if (err != 0)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto failure;
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_oif && fi->fib_nh->nh_oif != cfg->fc_oif)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto err_inval;
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_gw && fi->fib_nh->nh_gw != cfg->fc_gw)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto err_inval;
|
2011-01-14 05:36:42 -07:00
|
|
|
#ifdef CONFIG_IP_ROUTE_CLASSID
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_flow && fi->fib_nh->nh_tclassid != cfg->fc_flow)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto err_inval;
|
|
|
|
#endif
|
|
|
|
#else
|
|
|
|
goto err_inval;
|
|
|
|
#endif
|
|
|
|
} else {
|
|
|
|
struct fib_nh *nh = fi->fib_nh;
|
2006-08-17 19:14:52 -06:00
|
|
|
|
2015-07-21 02:43:47 -06:00
|
|
|
if (cfg->fc_encap) {
|
|
|
|
struct lwtunnel_state *lwtstate;
|
|
|
|
|
|
|
|
if (cfg->fc_encap_type == LWTUNNEL_ENCAP_NONE)
|
|
|
|
goto err_inval;
|
2017-01-30 13:07:37 -07:00
|
|
|
err = lwtunnel_build_state(cfg->fc_encap_type,
|
2015-08-24 10:45:41 -06:00
|
|
|
cfg->fc_encap, AF_INET, cfg,
|
|
|
|
&lwtstate);
|
2015-07-21 02:43:47 -06:00
|
|
|
if (err)
|
|
|
|
goto failure;
|
|
|
|
|
2015-07-24 04:28:36 -06:00
|
|
|
nh->nh_lwtstate = lwtstate_get(lwtstate);
|
2015-07-21 02:43:47 -06:00
|
|
|
}
|
2006-08-17 19:14:52 -06:00
|
|
|
nh->nh_oif = cfg->fc_oif;
|
|
|
|
nh->nh_gw = cfg->fc_gw;
|
|
|
|
nh->nh_flags = cfg->fc_flags;
|
2011-01-14 05:36:42 -07:00
|
|
|
#ifdef CONFIG_IP_ROUTE_CLASSID
|
2006-08-17 19:14:52 -06:00
|
|
|
nh->nh_tclassid = cfg->fc_flow;
|
2012-06-29 02:32:45 -06:00
|
|
|
if (nh->nh_tclassid)
|
2012-07-05 23:13:13 -06:00
|
|
|
fi->fib_net->ipv4.fib_num_tclassid_users++;
|
2005-04-16 16:20:36 -06:00
|
|
|
#endif
|
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
|
|
|
nh->nh_weight = 1;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
if (fib_props[cfg->fc_type].error) {
|
|
|
|
if (cfg->fc_gw || cfg->fc_oif || cfg->fc_mp)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto err_inval;
|
|
|
|
goto link_it;
|
2011-03-07 15:27:38 -07:00
|
|
|
} else {
|
|
|
|
switch (cfg->fc_type) {
|
|
|
|
case RTN_UNICAST:
|
|
|
|
case RTN_LOCAL:
|
|
|
|
case RTN_BROADCAST:
|
|
|
|
case RTN_ANYCAST:
|
|
|
|
case RTN_MULTICAST:
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
goto err_inval;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_scope > RT_SCOPE_HOST)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto err_inval;
|
|
|
|
|
2006-08-17 19:14:52 -06:00
|
|
|
if (cfg->fc_scope == RT_SCOPE_HOST) {
|
2005-04-16 16:20:36 -06:00
|
|
|
struct fib_nh *nh = fi->fib_nh;
|
|
|
|
|
|
|
|
/* Local address is added. */
|
|
|
|
if (nhs != 1 || nh->nh_gw)
|
|
|
|
goto err_inval;
|
|
|
|
nh->nh_scope = RT_SCOPE_NOWHERE;
|
2008-01-31 19:49:32 -07:00
|
|
|
nh->nh_dev = dev_get_by_index(net, fi->fib_nh->nh_oif);
|
2005-04-16 16:20:36 -06:00
|
|
|
err = -ENODEV;
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!nh->nh_dev)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto failure;
|
|
|
|
} else {
|
2015-06-23 11:45:36 -06:00
|
|
|
int linkdown = 0;
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
change_nexthops(fi) {
|
2010-10-04 14:00:18 -06:00
|
|
|
err = fib_check_nh(cfg, fi, nexthop_nh);
|
|
|
|
if (err != 0)
|
2005-04-16 16:20:36 -06:00
|
|
|
goto failure;
|
2015-06-23 11:45:36 -06:00
|
|
|
if (nexthop_nh->nh_flags & RTNH_F_LINKDOWN)
|
|
|
|
linkdown++;
|
2005-04-16 16:20:36 -06:00
|
|
|
} endfor_nexthops(fi)
|
2015-06-23 11:45:36 -06:00
|
|
|
if (linkdown == fi->fib_nhs)
|
|
|
|
fi->fib_flags |= RTNH_F_LINKDOWN;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2015-08-13 14:59:06 -06:00
|
|
|
if (fi->fib_prefsrc && !fib_valid_prefsrc(cfg, fi->fib_prefsrc))
|
|
|
|
goto err_inval;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-03-07 21:54:48 -07:00
|
|
|
change_nexthops(fi) {
|
2011-03-24 18:42:21 -06:00
|
|
|
fib_info_update_nh_saddr(net, nexthop_nh);
|
2015-09-30 02:12:21 -06:00
|
|
|
fib_add_weight(fi, nexthop_nh);
|
2011-03-07 21:54:48 -07:00
|
|
|
} endfor_nexthops(fi)
|
|
|
|
|
2015-09-30 02:12:21 -06:00
|
|
|
fib_rebalance(fi);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
link_it:
|
2010-10-04 14:00:18 -06:00
|
|
|
ofi = fib_find_info(fi);
|
|
|
|
if (ofi) {
|
2005-04-16 16:20:36 -06:00
|
|
|
fi->fib_dead = 1;
|
|
|
|
free_fib_info(fi);
|
|
|
|
ofi->fib_treeref++;
|
|
|
|
return ofi;
|
|
|
|
}
|
|
|
|
|
|
|
|
fi->fib_treeref++;
|
|
|
|
atomic_inc(&fi->fib_clntref);
|
2006-08-29 17:48:09 -06:00
|
|
|
spin_lock_bh(&fib_info_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
hlist_add_head(&fi->fib_hash,
|
|
|
|
&fib_info_hash[fib_info_hashfn(fi)]);
|
|
|
|
if (fi->fib_prefsrc) {
|
|
|
|
struct hlist_head *head;
|
|
|
|
|
|
|
|
head = &fib_info_laddrhash[fib_laddr_hashfn(fi->fib_prefsrc)];
|
|
|
|
hlist_add_head(&fi->fib_lhash, head);
|
|
|
|
}
|
|
|
|
change_nexthops(fi) {
|
|
|
|
struct hlist_head *head;
|
|
|
|
unsigned int hash;
|
|
|
|
|
2010-01-15 02:16:40 -07:00
|
|
|
if (!nexthop_nh->nh_dev)
|
2005-04-16 16:20:36 -06:00
|
|
|
continue;
|
2010-01-15 02:16:40 -07:00
|
|
|
hash = fib_devindex_hashfn(nexthop_nh->nh_dev->ifindex);
|
2005-04-16 16:20:36 -06:00
|
|
|
head = &fib_info_devhash[hash];
|
2010-01-15 02:16:40 -07:00
|
|
|
hlist_add_head(&nexthop_nh->nh_hash, head);
|
2005-04-16 16:20:36 -06:00
|
|
|
} endfor_nexthops(fi)
|
2006-08-29 17:48:09 -06:00
|
|
|
spin_unlock_bh(&fib_info_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
return fi;
|
|
|
|
|
|
|
|
err_inval:
|
|
|
|
err = -EINVAL;
|
|
|
|
|
|
|
|
failure:
|
2007-02-09 07:24:47 -07:00
|
|
|
if (fi) {
|
2005-04-16 16:20:36 -06:00
|
|
|
fi->fib_dead = 1;
|
|
|
|
free_fib_info(fi);
|
|
|
|
}
|
2006-08-17 19:14:52 -06:00
|
|
|
|
|
|
|
return ERR_PTR(err);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2012-09-07 14:12:54 -06:00
|
|
|
int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
|
2011-03-24 19:06:47 -06:00
|
|
|
u32 tb_id, u8 type, __be32 dst, int dst_len, u8 tos,
|
2006-08-17 19:15:17 -06:00
|
|
|
struct fib_info *fi, unsigned int flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-08-17 19:15:17 -06:00
|
|
|
struct nlmsghdr *nlh;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct rtmsg *rtm;
|
|
|
|
|
2012-09-07 14:12:54 -06:00
|
|
|
nlh = nlmsg_put(skb, portid, seq, event, sizeof(*rtm), flags);
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!nlh)
|
2007-02-01 00:16:40 -07:00
|
|
|
return -EMSGSIZE;
|
2006-08-17 19:15:17 -06:00
|
|
|
|
|
|
|
rtm = nlmsg_data(nlh);
|
2005-04-16 16:20:36 -06:00
|
|
|
rtm->rtm_family = AF_INET;
|
|
|
|
rtm->rtm_dst_len = dst_len;
|
|
|
|
rtm->rtm_src_len = 0;
|
|
|
|
rtm->rtm_tos = tos;
|
2008-06-10 16:44:49 -06:00
|
|
|
if (tb_id < 256)
|
|
|
|
rtm->rtm_table = tb_id;
|
|
|
|
else
|
|
|
|
rtm->rtm_table = RT_TABLE_COMPAT;
|
2012-04-01 18:39:02 -06:00
|
|
|
if (nla_put_u32(skb, RTA_TABLE, tb_id))
|
|
|
|
goto nla_put_failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
rtm->rtm_type = type;
|
|
|
|
rtm->rtm_flags = fi->fib_flags;
|
2011-03-24 19:06:47 -06:00
|
|
|
rtm->rtm_scope = fi->fib_scope;
|
2005-04-16 16:20:36 -06:00
|
|
|
rtm->rtm_protocol = fi->fib_protocol;
|
2006-08-17 19:15:17 -06:00
|
|
|
|
2012-04-01 18:39:02 -06:00
|
|
|
if (rtm->rtm_dst_len &&
|
2015-03-29 08:59:25 -06:00
|
|
|
nla_put_in_addr(skb, RTA_DST, dst))
|
2012-04-01 18:39:02 -06:00
|
|
|
goto nla_put_failure;
|
|
|
|
if (fi->fib_priority &&
|
|
|
|
nla_put_u32(skb, RTA_PRIORITY, fi->fib_priority))
|
|
|
|
goto nla_put_failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
if (rtnetlink_put_metrics(skb, fi->fib_metrics) < 0)
|
2006-08-17 19:15:17 -06:00
|
|
|
goto nla_put_failure;
|
|
|
|
|
2012-04-01 18:39:02 -06:00
|
|
|
if (fi->fib_prefsrc &&
|
2015-03-29 08:59:25 -06:00
|
|
|
nla_put_in_addr(skb, RTA_PREFSRC, fi->fib_prefsrc))
|
2012-04-01 18:39:02 -06:00
|
|
|
goto nla_put_failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
if (fi->fib_nhs == 1) {
|
2015-06-23 11:45:37 -06:00
|
|
|
struct in_device *in_dev;
|
|
|
|
|
2012-04-01 18:39:02 -06:00
|
|
|
if (fi->fib_nh->nh_gw &&
|
2015-03-29 08:59:25 -06:00
|
|
|
nla_put_in_addr(skb, RTA_GATEWAY, fi->fib_nh->nh_gw))
|
2012-04-01 18:39:02 -06:00
|
|
|
goto nla_put_failure;
|
|
|
|
if (fi->fib_nh->nh_oif &&
|
|
|
|
nla_put_u32(skb, RTA_OIF, fi->fib_nh->nh_oif))
|
|
|
|
goto nla_put_failure;
|
2015-06-23 11:45:37 -06:00
|
|
|
if (fi->fib_nh->nh_flags & RTNH_F_LINKDOWN) {
|
2015-06-26 17:37:11 -06:00
|
|
|
in_dev = __in_dev_get_rtnl(fi->fib_nh->nh_dev);
|
2015-06-23 11:45:37 -06:00
|
|
|
if (in_dev &&
|
|
|
|
IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev))
|
|
|
|
rtm->rtm_flags |= RTNH_F_DEAD;
|
|
|
|
}
|
2011-01-14 05:36:42 -07:00
|
|
|
#ifdef CONFIG_IP_ROUTE_CLASSID
|
2012-04-01 18:39:02 -06:00
|
|
|
if (fi->fib_nh[0].nh_tclassid &&
|
|
|
|
nla_put_u32(skb, RTA_FLOW, fi->fib_nh[0].nh_tclassid))
|
|
|
|
goto nla_put_failure;
|
2006-07-21 16:09:55 -06:00
|
|
|
#endif
|
2017-01-11 15:29:54 -07:00
|
|
|
if (fi->fib_nh->nh_lwtstate &&
|
|
|
|
lwtunnel_fill_encap(skb, fi->fib_nh->nh_lwtstate) < 0)
|
|
|
|
goto nla_put_failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
|
|
|
if (fi->fib_nhs > 1) {
|
2006-08-17 19:15:17 -06:00
|
|
|
struct rtnexthop *rtnh;
|
|
|
|
struct nlattr *mp;
|
|
|
|
|
|
|
|
mp = nla_nest_start(skb, RTA_MULTIPATH);
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!mp)
|
2006-08-17 19:15:17 -06:00
|
|
|
goto nla_put_failure;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
for_nexthops(fi) {
|
2015-06-23 11:45:37 -06:00
|
|
|
struct in_device *in_dev;
|
|
|
|
|
2006-08-17 19:15:17 -06:00
|
|
|
rtnh = nla_reserve_nohdr(skb, sizeof(*rtnh));
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!rtnh)
|
2006-08-17 19:15:17 -06:00
|
|
|
goto nla_put_failure;
|
|
|
|
|
|
|
|
rtnh->rtnh_flags = nh->nh_flags & 0xFF;
|
2015-06-23 11:45:37 -06:00
|
|
|
if (nh->nh_flags & RTNH_F_LINKDOWN) {
|
2015-06-26 17:37:11 -06:00
|
|
|
in_dev = __in_dev_get_rtnl(nh->nh_dev);
|
2015-06-23 11:45:37 -06:00
|
|
|
if (in_dev &&
|
|
|
|
IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev))
|
|
|
|
rtnh->rtnh_flags |= RTNH_F_DEAD;
|
|
|
|
}
|
2006-08-17 19:15:17 -06:00
|
|
|
rtnh->rtnh_hops = nh->nh_weight - 1;
|
|
|
|
rtnh->rtnh_ifindex = nh->nh_oif;
|
|
|
|
|
2012-04-01 18:39:02 -06:00
|
|
|
if (nh->nh_gw &&
|
2015-03-29 08:59:25 -06:00
|
|
|
nla_put_in_addr(skb, RTA_GATEWAY, nh->nh_gw))
|
2012-04-01 18:39:02 -06:00
|
|
|
goto nla_put_failure;
|
2011-01-14 05:36:42 -07:00
|
|
|
#ifdef CONFIG_IP_ROUTE_CLASSID
|
2012-04-01 18:39:02 -06:00
|
|
|
if (nh->nh_tclassid &&
|
|
|
|
nla_put_u32(skb, RTA_FLOW, nh->nh_tclassid))
|
|
|
|
goto nla_put_failure;
|
2006-07-21 16:09:55 -06:00
|
|
|
#endif
|
2017-01-11 15:29:54 -07:00
|
|
|
if (nh->nh_lwtstate &&
|
|
|
|
lwtunnel_fill_encap(skb, nh->nh_lwtstate) < 0)
|
|
|
|
goto nla_put_failure;
|
|
|
|
|
2006-08-17 19:15:17 -06:00
|
|
|
/* length of rtnetlink header + attributes */
|
|
|
|
rtnh->rtnh_len = nlmsg_get_pos(skb) - (void *) rtnh;
|
2005-04-16 16:20:36 -06:00
|
|
|
} endfor_nexthops(fi);
|
2006-08-17 19:15:17 -06:00
|
|
|
|
|
|
|
nla_nest_end(skb, mp);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
#endif
|
2015-01-16 14:09:00 -07:00
|
|
|
nlmsg_end(skb, nlh);
|
|
|
|
return 0;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-08-17 19:15:17 -06:00
|
|
|
nla_put_failure:
|
2007-02-01 00:16:40 -07:00
|
|
|
nlmsg_cancel(skb, nlh);
|
|
|
|
return -EMSGSIZE;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2010-10-04 14:00:18 -06:00
|
|
|
* Update FIB if:
|
|
|
|
* - local address disappeared -> we must delete all the entries
|
|
|
|
* referring to it.
|
|
|
|
* - device went down -> we must shutdown all nexthops going via it.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2016-09-04 16:20:20 -06:00
|
|
|
int fib_sync_down_addr(struct net_device *dev, __be32 local)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
int ret = 0;
|
2008-01-31 19:48:47 -07:00
|
|
|
unsigned int hash = fib_laddr_hashfn(local);
|
|
|
|
struct hlist_head *head = &fib_info_laddrhash[hash];
|
2016-09-04 16:20:20 -06:00
|
|
|
struct net *net = dev_net(dev);
|
|
|
|
int tb_id = l3mdev_fib_table(dev);
|
2008-01-31 19:48:47 -07:00
|
|
|
struct fib_info *fi;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!fib_info_laddrhash || local == 0)
|
2008-01-31 19:48:47 -07:00
|
|
|
return 0;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-27 18:06:00 -07:00
|
|
|
hlist_for_each_entry(fi, head, fib_lhash) {
|
2016-09-04 16:20:20 -06:00
|
|
|
if (!net_eq(fi->fib_net, net) ||
|
|
|
|
fi->fib_tb_id != tb_id)
|
2008-01-31 19:50:07 -07:00
|
|
|
continue;
|
2008-01-31 19:48:47 -07:00
|
|
|
if (fi->fib_prefsrc == local) {
|
|
|
|
fi->fib_flags |= RTNH_F_DEAD;
|
|
|
|
ret++;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
}
|
2008-01-31 19:48:47 -07:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2015-10-30 02:23:33 -06:00
|
|
|
/* Event force Flags Description
|
|
|
|
* NETDEV_CHANGE 0 LINKDOWN Carrier OFF, not for scope host
|
|
|
|
* NETDEV_DOWN 0 LINKDOWN|DEAD Link down, not for scope host
|
|
|
|
* NETDEV_DOWN 1 LINKDOWN|DEAD Last address removed
|
|
|
|
* NETDEV_UNREGISTER 1 LINKDOWN|DEAD Device removed
|
|
|
|
*/
|
|
|
|
int fib_sync_down_dev(struct net_device *dev, unsigned long event, bool force)
|
2008-01-31 19:48:47 -07:00
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
int scope = RT_SCOPE_NOWHERE;
|
|
|
|
struct fib_info *prev_fi = NULL;
|
|
|
|
unsigned int hash = fib_devindex_hashfn(dev->ifindex);
|
|
|
|
struct hlist_head *head = &fib_info_devhash[hash];
|
|
|
|
struct fib_nh *nh;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2015-10-30 02:23:33 -06:00
|
|
|
if (force)
|
2008-01-31 19:48:47 -07:00
|
|
|
scope = -1;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-27 18:06:00 -07:00
|
|
|
hlist_for_each_entry(nh, head, nh_hash) {
|
2008-01-31 19:48:47 -07:00
|
|
|
struct fib_info *fi = nh->nh_parent;
|
|
|
|
int dead;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-01-31 19:48:47 -07:00
|
|
|
BUG_ON(!fi->fib_nhs);
|
|
|
|
if (nh->nh_dev != dev || fi == prev_fi)
|
|
|
|
continue;
|
|
|
|
prev_fi = fi;
|
|
|
|
dead = 0;
|
|
|
|
change_nexthops(fi) {
|
2010-10-04 14:00:18 -06:00
|
|
|
if (nexthop_nh->nh_flags & RTNH_F_DEAD)
|
2008-01-31 19:48:47 -07:00
|
|
|
dead++;
|
2010-01-15 02:16:40 -07:00
|
|
|
else if (nexthop_nh->nh_dev == dev &&
|
|
|
|
nexthop_nh->nh_scope != scope) {
|
2015-06-23 11:45:36 -06:00
|
|
|
switch (event) {
|
|
|
|
case NETDEV_DOWN:
|
|
|
|
case NETDEV_UNREGISTER:
|
|
|
|
nexthop_nh->nh_flags |= RTNH_F_DEAD;
|
|
|
|
/* fall through */
|
|
|
|
case NETDEV_CHANGE:
|
|
|
|
nexthop_nh->nh_flags |= RTNH_F_LINKDOWN;
|
|
|
|
break;
|
|
|
|
}
|
2008-01-31 19:48:47 -07:00
|
|
|
dead++;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
2015-06-23 11:45:36 -06:00
|
|
|
if (event == NETDEV_UNREGISTER &&
|
|
|
|
nexthop_nh->nh_dev == dev) {
|
2008-01-31 19:48:47 -07:00
|
|
|
dead = fi->fib_nhs;
|
|
|
|
break;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2008-01-31 19:48:47 -07:00
|
|
|
#endif
|
|
|
|
} endfor_nexthops(fi)
|
|
|
|
if (dead == fi->fib_nhs) {
|
2015-06-23 11:45:36 -06:00
|
|
|
switch (event) {
|
|
|
|
case NETDEV_DOWN:
|
|
|
|
case NETDEV_UNREGISTER:
|
|
|
|
fi->fib_flags |= RTNH_F_DEAD;
|
|
|
|
/* fall through */
|
|
|
|
case NETDEV_CHANGE:
|
|
|
|
fi->fib_flags |= RTNH_F_LINKDOWN;
|
|
|
|
break;
|
|
|
|
}
|
2008-01-31 19:48:47 -07:00
|
|
|
ret++;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2015-09-30 02:12:21 -06:00
|
|
|
|
|
|
|
fib_rebalance(fi);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2011-01-31 17:16:50 -07:00
|
|
|
/* Must be invoked inside of an RCU protected region. */
|
2017-01-05 20:33:59 -07:00
|
|
|
static void fib_select_default(const struct flowi4 *flp, struct fib_result *res)
|
2011-01-31 17:16:50 -07:00
|
|
|
{
|
|
|
|
struct fib_info *fi = NULL, *last_resort = NULL;
|
2015-02-25 16:31:31 -07:00
|
|
|
struct hlist_head *fa_head = res->fa_head;
|
2011-01-31 17:16:50 -07:00
|
|
|
struct fib_table *tb = res->table;
|
2015-07-22 01:43:22 -06:00
|
|
|
u8 slen = 32 - res->prefixlen;
|
2011-01-31 17:16:50 -07:00
|
|
|
int order = -1, last_idx = -1;
|
2015-07-22 01:43:23 -06:00
|
|
|
struct fib_alias *fa, *fa1 = NULL;
|
|
|
|
u32 last_prio = res->fi->fib_priority;
|
|
|
|
u8 last_tos = 0;
|
2011-01-31 17:16:50 -07:00
|
|
|
|
2015-02-25 16:31:31 -07:00
|
|
|
hlist_for_each_entry_rcu(fa, fa_head, fa_list) {
|
2011-01-31 17:16:50 -07:00
|
|
|
struct fib_info *next_fi = fa->fa_info;
|
|
|
|
|
2015-07-22 01:43:22 -06:00
|
|
|
if (fa->fa_slen != slen)
|
|
|
|
continue;
|
2015-07-22 01:43:23 -06:00
|
|
|
if (fa->fa_tos && fa->fa_tos != flp->flowi4_tos)
|
|
|
|
continue;
|
2015-07-22 01:43:22 -06:00
|
|
|
if (fa->tb_id != tb->tb_id)
|
|
|
|
continue;
|
2015-07-22 01:43:23 -06:00
|
|
|
if (next_fi->fib_priority > last_prio &&
|
|
|
|
fa->fa_tos == last_tos) {
|
|
|
|
if (last_tos)
|
|
|
|
continue;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (next_fi->fib_flags & RTNH_F_DEAD)
|
|
|
|
continue;
|
|
|
|
last_tos = fa->fa_tos;
|
|
|
|
last_prio = next_fi->fib_priority;
|
|
|
|
|
2011-03-24 19:06:47 -06:00
|
|
|
if (next_fi->fib_scope != res->scope ||
|
2011-01-31 17:16:50 -07:00
|
|
|
fa->fa_type != RTN_UNICAST)
|
|
|
|
continue;
|
|
|
|
if (!next_fi->fib_nh[0].nh_gw ||
|
|
|
|
next_fi->fib_nh[0].nh_scope != RT_SCOPE_LINK)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
fib_alias_accessed(fa);
|
|
|
|
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!fi) {
|
2011-01-31 17:16:50 -07:00
|
|
|
if (next_fi != res->fi)
|
|
|
|
break;
|
2015-07-22 01:43:23 -06:00
|
|
|
fa1 = fa;
|
2011-01-31 17:16:50 -07:00
|
|
|
} else if (!fib_detect_death(fi, order, &last_resort,
|
2015-07-22 01:43:23 -06:00
|
|
|
&last_idx, fa1->fa_default)) {
|
2011-01-31 17:16:50 -07:00
|
|
|
fib_result_assign(res, fi);
|
2015-07-22 01:43:23 -06:00
|
|
|
fa1->fa_default = order;
|
2011-01-31 17:16:50 -07:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
fi = next_fi;
|
|
|
|
order++;
|
|
|
|
}
|
|
|
|
|
2015-04-03 02:17:26 -06:00
|
|
|
if (order <= 0 || !fi) {
|
2015-07-22 01:43:23 -06:00
|
|
|
if (fa1)
|
|
|
|
fa1->fa_default = -1;
|
2011-01-31 17:16:50 -07:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!fib_detect_death(fi, order, &last_resort, &last_idx,
|
2015-07-22 01:43:23 -06:00
|
|
|
fa1->fa_default)) {
|
2011-01-31 17:16:50 -07:00
|
|
|
fib_result_assign(res, fi);
|
2015-07-22 01:43:23 -06:00
|
|
|
fa1->fa_default = order;
|
2011-01-31 17:16:50 -07:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (last_idx >= 0)
|
|
|
|
fib_result_assign(res, last_resort);
|
2015-07-22 01:43:23 -06:00
|
|
|
fa1->fa_default = last_idx;
|
2011-01-31 17:16:50 -07:00
|
|
|
out:
|
2011-02-14 12:23:04 -07:00
|
|
|
return;
|
2011-01-31 17:16:50 -07:00
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
2010-10-04 14:00:18 -06:00
|
|
|
* Dead device goes up. We wake up dead nexthops.
|
|
|
|
* It takes sense only on multipath routes.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2015-06-23 11:45:36 -06:00
|
|
|
int fib_sync_up(struct net_device *dev, unsigned int nh_flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct fib_info *prev_fi;
|
|
|
|
unsigned int hash;
|
|
|
|
struct hlist_head *head;
|
|
|
|
struct fib_nh *nh;
|
|
|
|
int ret;
|
|
|
|
|
2010-10-04 14:00:18 -06:00
|
|
|
if (!(dev->flags & IFF_UP))
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
|
2015-10-30 02:23:34 -06:00
|
|
|
if (nh_flags & RTNH_F_DEAD) {
|
|
|
|
unsigned int flags = dev_get_flags(dev);
|
|
|
|
|
|
|
|
if (flags & (IFF_RUNNING | IFF_LOWER_UP))
|
|
|
|
nh_flags |= RTNH_F_LINKDOWN;
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
prev_fi = NULL;
|
|
|
|
hash = fib_devindex_hashfn(dev->ifindex);
|
|
|
|
head = &fib_info_devhash[hash];
|
|
|
|
ret = 0;
|
|
|
|
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-27 18:06:00 -07:00
|
|
|
hlist_for_each_entry(nh, head, nh_hash) {
|
2005-04-16 16:20:36 -06:00
|
|
|
struct fib_info *fi = nh->nh_parent;
|
|
|
|
int alive;
|
|
|
|
|
|
|
|
BUG_ON(!fi->fib_nhs);
|
|
|
|
if (nh->nh_dev != dev || fi == prev_fi)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
prev_fi = fi;
|
|
|
|
alive = 0;
|
|
|
|
change_nexthops(fi) {
|
2015-06-23 11:45:36 -06:00
|
|
|
if (!(nexthop_nh->nh_flags & nh_flags)) {
|
2005-04-16 16:20:36 -06:00
|
|
|
alive++;
|
|
|
|
continue;
|
|
|
|
}
|
2015-04-03 02:17:26 -06:00
|
|
|
if (!nexthop_nh->nh_dev ||
|
2010-10-04 14:00:18 -06:00
|
|
|
!(nexthop_nh->nh_dev->flags & IFF_UP))
|
2005-04-16 16:20:36 -06:00
|
|
|
continue;
|
2010-01-15 02:16:40 -07:00
|
|
|
if (nexthop_nh->nh_dev != dev ||
|
|
|
|
!__in_dev_get_rtnl(dev))
|
2005-04-16 16:20:36 -06:00
|
|
|
continue;
|
|
|
|
alive++;
|
2015-06-23 11:45:36 -06:00
|
|
|
nexthop_nh->nh_flags &= ~nh_flags;
|
2005-04-16 16:20:36 -06:00
|
|
|
} endfor_nexthops(fi)
|
|
|
|
|
|
|
|
if (alive > 0) {
|
2015-06-23 11:45:36 -06:00
|
|
|
fi->fib_flags &= ~nh_flags;
|
2005-04-16 16:20:36 -06:00
|
|
|
ret++;
|
|
|
|
}
|
2015-09-30 02:12:21 -06:00
|
|
|
|
|
|
|
fib_rebalance(fi);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2015-06-23 11:45:36 -06:00
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
2016-04-07 08:21:00 -06:00
|
|
|
static bool fib_good_nh(const struct fib_nh *nh)
|
|
|
|
{
|
|
|
|
int state = NUD_REACHABLE;
|
|
|
|
|
|
|
|
if (nh->nh_scope == RT_SCOPE_LINK) {
|
|
|
|
struct neighbour *n;
|
|
|
|
|
|
|
|
rcu_read_lock_bh();
|
|
|
|
|
2016-08-18 11:19:34 -06:00
|
|
|
n = __ipv4_neigh_lookup_noref(nh->nh_dev,
|
|
|
|
(__force u32)nh->nh_gw);
|
2016-04-07 08:21:00 -06:00
|
|
|
if (n)
|
|
|
|
state = n->nud_state;
|
|
|
|
|
|
|
|
rcu_read_unlock_bh();
|
|
|
|
}
|
|
|
|
|
|
|
|
return !!(state & NUD_VALID);
|
|
|
|
}
|
2015-06-23 11:45:36 -06:00
|
|
|
|
2015-09-30 02:12:21 -06:00
|
|
|
void fib_select_multipath(struct fib_result *res, int hash)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct fib_info *fi = res->fi;
|
2016-04-07 08:21:00 -06:00
|
|
|
struct net *net = fi->fib_net;
|
|
|
|
bool first = false;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2015-09-30 02:12:21 -06:00
|
|
|
for_nexthops(fi) {
|
|
|
|
if (hash > atomic_read(&nh->nh_upper_bound))
|
|
|
|
continue;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2016-04-07 08:21:00 -06:00
|
|
|
if (!net->ipv4.sysctl_fib_multipath_use_neigh ||
|
|
|
|
fib_good_nh(nh)) {
|
|
|
|
res->nh_sel = nhsel;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
if (!first) {
|
|
|
|
res->nh_sel = nhsel;
|
|
|
|
first = true;
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
} endfor_nexthops(fi);
|
|
|
|
}
|
|
|
|
#endif
|
2015-10-05 09:51:25 -06:00
|
|
|
|
|
|
|
void fib_select_path(struct net *net, struct fib_result *res,
|
|
|
|
struct flowi4 *fl4, int mp_hash)
|
|
|
|
{
|
2017-01-10 15:37:35 -07:00
|
|
|
bool oif_check;
|
|
|
|
|
|
|
|
oif_check = (fl4->flowi4_oif == 0 ||
|
|
|
|
fl4->flowi4_flags & FLOWI_FLAG_SKIP_NH_OIF);
|
|
|
|
|
2015-10-05 09:51:25 -06:00
|
|
|
#ifdef CONFIG_IP_ROUTE_MULTIPATH
|
2017-01-10 15:37:35 -07:00
|
|
|
if (res->fi->fib_nhs > 1 && oif_check) {
|
2015-10-05 09:51:25 -06:00
|
|
|
if (mp_hash < 0)
|
2015-10-29 15:20:40 -06:00
|
|
|
mp_hash = get_hash_from_flowi4(fl4) >> 1;
|
|
|
|
|
2015-10-05 09:51:25 -06:00
|
|
|
fib_select_multipath(res, mp_hash);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
#endif
|
|
|
|
if (!res->prefixlen &&
|
|
|
|
res->table->tb_num_default > 1 &&
|
2017-01-10 15:37:35 -07:00
|
|
|
res->type == RTN_UNICAST && oif_check)
|
2015-10-05 09:51:25 -06:00
|
|
|
fib_select_default(fl4, res);
|
|
|
|
|
|
|
|
if (!fl4->saddr)
|
|
|
|
fl4->saddr = FIB_RES_PREFSRC(net, *res);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(fib_select_path);
|