2005-04-17 02:20:36 +04:00
/*
* 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 : FIB frontend .
*
* 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 .
*/
# include <linux/module.h>
2016-12-24 22:46:01 +03:00
# include <linux/uaccess.h>
2005-04-17 02:20:36 +04:00
# include <linux/bitops.h>
2006-01-11 23:17:47 +03:00
# include <linux/capability.h>
2005-04-17 02:20:36 +04:00
# include <linux/types.h>
# include <linux/kernel.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>
2005-12-27 07:43:12 +03:00
# include <linux/inetdevice.h>
2005-04-17 02:20:36 +04:00
# include <linux/netdevice.h>
2006-08-05 10:04:54 +04:00
# include <linux/if_addr.h>
2005-04-17 02:20:36 +04:00
# include <linux/if_arp.h>
# include <linux/skbuff.h>
2012-06-29 12:32:45 +04:00
# include <linux/cache.h>
2005-04-17 02:20:36 +04:00
# include <linux/init.h>
2006-08-11 10:10:46 +04:00
# include <linux/list.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 11:04:11 +03:00
# include <linux/slab.h>
2005-04-17 02:20:36 +04:00
# include <net/ip.h>
# include <net/protocol.h>
# include <net/route.h>
# include <net/tcp.h>
# include <net/sock.h>
# include <net/arp.h>
# include <net/ip_fib.h>
2007-03-22 21:55:17 +03:00
# include <net/rtnetlink.h>
2011-04-07 08:51:51 +04:00
# include <net/xfrm.h>
2015-09-30 06:07:13 +03:00
# include <net/l3mdev.h>
2017-01-18 01:57:36 +03:00
# include <net/lwtunnel.h>
2015-08-28 18:42:09 +03:00
# include <trace/events/fib.h>
2005-04-17 02:20:36 +04:00
# ifndef CONFIG_IP_MULTIPLE_TABLES
2008-01-10 14:22:17 +03:00
static int __net_init fib4_rules_init ( struct net * net )
2007-11-07 10:34:04 +03:00
{
2008-01-10 14:23:38 +03:00
struct fib_table * local_table , * main_table ;
2015-03-07 00:47:00 +03:00
main_table = fib_trie_table ( RT_TABLE_MAIN , NULL ) ;
2015-04-03 11:17:26 +03:00
if ( ! main_table )
2015-03-12 00:02:16 +03:00
return - ENOMEM ;
2008-01-10 14:21:49 +03:00
2015-03-07 00:47:00 +03:00
local_table = fib_trie_table ( RT_TABLE_LOCAL , main_table ) ;
2015-04-03 11:17:26 +03:00
if ( ! local_table )
2015-03-12 00:02:16 +03:00
goto fail ;
2015-03-07 00:47:00 +03:00
2008-01-10 14:23:38 +03:00
hlist_add_head_rcu ( & local_table - > tb_hlist ,
2008-01-10 14:28:24 +03:00
& net - > ipv4 . fib_table_hash [ TABLE_LOCAL_INDEX ] ) ;
2008-01-10 14:23:38 +03:00
hlist_add_head_rcu ( & main_table - > tb_hlist ,
2008-01-10 14:28:24 +03:00
& net - > ipv4 . fib_table_hash [ TABLE_MAIN_INDEX ] ) ;
2008-01-10 14:21:49 +03:00
return 0 ;
fail :
2015-03-12 00:02:16 +03:00
fib_free_table ( main_table ) ;
2008-01-10 14:21:49 +03:00
return - ENOMEM ;
2007-11-07 10:34:04 +03:00
}
2006-08-11 10:10:46 +04:00
# else
2005-04-17 02:20:36 +04:00
2008-01-10 14:24:11 +03:00
struct fib_table * fib_new_table ( struct net * net , u32 id )
2005-04-17 02:20:36 +04:00
{
2015-03-07 00:47:00 +03:00
struct fib_table * tb , * alias = NULL ;
2006-08-11 10:10:46 +04:00
unsigned int h ;
2005-04-17 02:20:36 +04:00
2006-08-11 10:10:46 +04:00
if ( id = = 0 )
id = RT_TABLE_MAIN ;
2008-01-10 14:24:11 +03:00
tb = fib_get_table ( net , id ) ;
2006-08-11 10:10:46 +04:00
if ( tb )
return tb ;
2008-01-15 10:14:20 +03:00
2017-01-03 00:32:54 +03:00
if ( id = = RT_TABLE_LOCAL & & ! net - > ipv4 . fib_has_custom_rules )
2015-03-07 00:47:00 +03:00
alias = fib_new_table ( net , RT_TABLE_MAIN ) ;
tb = fib_trie_table ( id , alias ) ;
2005-04-17 02:20:36 +04:00
if ( ! tb )
return NULL ;
2012-07-06 09:13:13 +04:00
switch ( id ) {
case RT_TABLE_MAIN :
2015-03-05 02:02:44 +03:00
rcu_assign_pointer ( net - > ipv4 . fib_main , tb ) ;
2012-07-06 09:13:13 +04:00
break ;
case RT_TABLE_DEFAULT :
2015-03-05 02:02:44 +03:00
rcu_assign_pointer ( net - > ipv4 . fib_default , tb ) ;
2012-07-06 09:13:13 +04:00
break ;
default :
break ;
}
2006-08-11 10:10:46 +04:00
h = id & ( FIB_TABLE_HASHSZ - 1 ) ;
2008-01-10 14:28:24 +03:00
hlist_add_head_rcu ( & tb - > tb_hlist , & net - > ipv4 . fib_table_hash [ h ] ) ;
2005-04-17 02:20:36 +04:00
return tb ;
}
2016-05-05 07:46:12 +03:00
EXPORT_SYMBOL_GPL ( fib_new_table ) ;
2005-04-17 02:20:36 +04:00
2014-12-31 21:56:24 +03:00
/* caller must hold either rtnl or rcu read lock */
2008-01-10 14:24:11 +03:00
struct fib_table * fib_get_table ( struct net * net , u32 id )
2006-08-11 10:10:46 +04:00
{
struct fib_table * tb ;
2008-01-10 14:28:24 +03:00
struct hlist_head * head ;
2006-08-11 10:10:46 +04:00
unsigned int h ;
2005-04-17 02:20:36 +04:00
2006-08-11 10:10:46 +04:00
if ( id = = 0 )
id = RT_TABLE_MAIN ;
h = id & ( FIB_TABLE_HASHSZ - 1 ) ;
2008-01-10 14:28:24 +03:00
head = & net - > ipv4 . fib_table_hash [ h ] ;
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>
2013-02-28 05:06:00 +04:00
hlist_for_each_entry_rcu ( tb , head , tb_hlist ) {
2014-12-31 21:56:24 +03:00
if ( tb - > tb_id = = id )
2006-08-11 10:10:46 +04:00
return tb ;
}
return NULL ;
}
2005-04-17 02:20:36 +04:00
# endif /* CONFIG_IP_MULTIPLE_TABLES */
2015-03-07 00:47:00 +03:00
static void fib_replace_table ( struct net * net , struct fib_table * old ,
struct fib_table * new )
{
# ifdef CONFIG_IP_MULTIPLE_TABLES
switch ( new - > tb_id ) {
case RT_TABLE_MAIN :
rcu_assign_pointer ( net - > ipv4 . fib_main , new ) ;
break ;
case RT_TABLE_DEFAULT :
rcu_assign_pointer ( net - > ipv4 . fib_default , new ) ;
break ;
default :
break ;
}
# endif
/* replace the old table in the hlist */
hlist_replace_rcu ( & old - > tb_hlist , & new - > tb_hlist ) ;
}
int fib_unmerge ( struct net * net )
{
2016-11-15 13:46:06 +03:00
struct fib_table * old , * new , * main_table ;
2015-03-07 00:47:00 +03:00
2015-03-13 00:46:23 +03:00
/* attempt to fetch local table if it has been allocated */
2015-03-07 00:47:00 +03:00
old = fib_get_table ( net , RT_TABLE_LOCAL ) ;
2015-03-13 00:46:23 +03:00
if ( ! old )
return 0 ;
2015-03-07 00:47:00 +03:00
2015-03-13 00:46:23 +03:00
new = fib_trie_unmerge ( old ) ;
2015-03-07 00:47:00 +03:00
if ( ! new )
return - ENOMEM ;
2016-11-15 13:46:06 +03:00
/* table is already unmerged */
if ( new = = old )
return 0 ;
2015-03-07 00:47:00 +03:00
/* replace merged table with clean table */
2016-11-15 13:46:06 +03:00
fib_replace_table ( net , old , new ) ;
fib_free_table ( old ) ;
/* attempt to fetch main table if it has been allocated */
main_table = fib_get_table ( net , RT_TABLE_MAIN ) ;
if ( ! main_table )
return 0 ;
/* flush local entries from main table */
fib_table_flush_external ( main_table ) ;
2015-03-07 00:47:00 +03:00
return 0 ;
}
2008-01-10 14:28:24 +03:00
static void fib_flush ( struct net * net )
2005-04-17 02:20:36 +04:00
{
int flushed = 0 ;
2006-08-11 10:10:46 +04:00
unsigned int h ;
2005-04-17 02:20:36 +04:00
2006-08-11 10:10:46 +04:00
for ( h = 0 ; h < FIB_TABLE_HASHSZ ; h + + ) {
2015-03-05 02:02:44 +03:00
struct hlist_head * head = & net - > ipv4 . fib_table_hash [ h ] ;
struct hlist_node * tmp ;
struct fib_table * tb ;
hlist_for_each_entry_safe ( tb , tmp , head , tb_hlist )
2016-09-26 13:52:29 +03:00
flushed + = fib_table_flush ( net , tb ) ;
2005-04-17 02:20:36 +04:00
}
if ( flushed )
2012-09-07 04:45:29 +04:00
rt_cache_flush ( net ) ;
2005-04-17 02:20:36 +04:00
}
2007-12-05 10:28:46 +03:00
/*
* Find address type as if only " dev " was present in the system . If
* on_dev is NULL then all interfaces are taken into consideration .
*/
2012-04-15 09:58:06 +04:00
static inline unsigned int __inet_dev_addr_type ( struct net * net ,
const struct net_device * dev ,
2015-09-01 23:26:35 +03:00
__be32 addr , u32 tb_id )
2005-04-17 02:20:36 +04:00
{
2011-03-12 10:02:42 +03:00
struct flowi4 fl4 = { . daddr = addr } ;
2005-04-17 02:20:36 +04:00
struct fib_result res ;
2012-04-15 09:58:06 +04:00
unsigned int ret = RTN_BROADCAST ;
2015-08-13 23:59:04 +03:00
struct fib_table * table ;
2005-04-17 02:20:36 +04:00
2008-01-21 14:18:08 +03:00
if ( ipv4_is_zeronet ( addr ) | | ipv4_is_lbcast ( addr ) )
2005-04-17 02:20:36 +04:00
return RTN_BROADCAST ;
2007-12-17 00:45:43 +03:00
if ( ipv4_is_multicast ( addr ) )
2005-04-17 02:20:36 +04:00
return RTN_MULTICAST ;
2014-12-31 21:56:24 +03:00
rcu_read_lock ( ) ;
2015-08-13 23:59:04 +03:00
table = fib_get_table ( net , tb_id ) ;
if ( table ) {
2005-04-17 02:20:36 +04:00
ret = RTN_UNICAST ;
2015-08-13 23:59:04 +03:00
if ( ! fib_table_lookup ( table , & fl4 , & res , FIB_LOOKUP_NOREF ) ) {
2007-12-05 10:28:46 +03:00
if ( ! dev | | dev = = res . fi - > fib_dev )
ret = res . type ;
2005-04-17 02:20:36 +04:00
}
}
2014-12-31 21:56:24 +03:00
rcu_read_unlock ( ) ;
2005-04-17 02:20:36 +04:00
return ret ;
}
2015-09-01 23:26:35 +03:00
unsigned int inet_addr_type_table ( struct net * net , __be32 addr , u32 tb_id )
2015-08-13 23:59:04 +03:00
{
return __inet_dev_addr_type ( net , NULL , addr , tb_id ) ;
}
EXPORT_SYMBOL ( inet_addr_type_table ) ;
2008-01-10 14:25:28 +03:00
unsigned int inet_addr_type ( struct net * net , __be32 addr )
2007-12-05 10:28:46 +03:00
{
2015-08-13 23:59:04 +03:00
return __inet_dev_addr_type ( net , NULL , addr , RT_TABLE_LOCAL ) ;
2007-12-05 10:28:46 +03:00
}
2010-07-10 01:22:10 +04:00
EXPORT_SYMBOL ( inet_addr_type ) ;
2007-12-05 10:28:46 +03:00
2008-01-10 14:25:28 +03:00
unsigned int inet_dev_addr_type ( struct net * net , const struct net_device * dev ,
__be32 addr )
2007-12-05 10:28:46 +03:00
{
2015-09-30 06:07:14 +03:00
u32 rt_table = l3mdev_fib_table ( dev ) ? : RT_TABLE_LOCAL ;
2015-08-13 23:59:04 +03:00
return __inet_dev_addr_type ( net , dev , addr , rt_table ) ;
2007-12-05 10:28:46 +03:00
}
2010-07-10 01:22:10 +04:00
EXPORT_SYMBOL ( inet_dev_addr_type ) ;
2007-12-05 10:28:46 +03:00
2015-08-13 23:59:05 +03:00
/* inet_addr_type with dev == NULL but using the table from a dev
* if one is associated
*/
unsigned int inet_addr_type_dev_table ( struct net * net ,
const struct net_device * dev ,
__be32 addr )
{
2015-09-30 06:07:14 +03:00
u32 rt_table = l3mdev_fib_table ( dev ) ? : RT_TABLE_LOCAL ;
2015-08-13 23:59:05 +03:00
return __inet_dev_addr_type ( net , NULL , addr , rt_table ) ;
}
EXPORT_SYMBOL ( inet_addr_type_dev_table ) ;
2012-06-28 14:59:11 +04:00
__be32 fib_compute_spec_dst ( struct sk_buff * skb )
{
struct net_device * dev = skb - > dev ;
struct in_device * in_dev ;
struct fib_result res ;
2012-06-29 05:33:24 +04:00
struct rtable * rt ;
2012-06-28 14:59:11 +04:00
struct net * net ;
2012-06-29 05:33:24 +04:00
int scope ;
2012-06-28 14:59:11 +04:00
2012-06-29 05:33:24 +04:00
rt = skb_rtable ( skb ) ;
2012-07-19 01:35:03 +04:00
if ( ( rt - > rt_flags & ( RTCF_BROADCAST | RTCF_MULTICAST | RTCF_LOCAL ) ) = =
RTCF_LOCAL )
2012-06-28 14:59:11 +04:00
return ip_hdr ( skb ) - > daddr ;
in_dev = __in_dev_get_rcu ( dev ) ;
BUG_ON ( ! in_dev ) ;
net = dev_net ( dev ) ;
2012-06-29 05:33:24 +04:00
scope = RT_SCOPE_UNIVERSE ;
if ( ! ipv4_is_zeronet ( ip_hdr ( skb ) - > saddr ) ) {
2016-03-22 21:56:57 +03:00
struct flowi4 fl4 = {
. flowi4_iif = LOOPBACK_IFINDEX ,
. daddr = ip_hdr ( skb ) - > saddr ,
. flowi4_tos = RT_TOS ( ip_hdr ( skb ) - > tos ) ,
. flowi4_scope = scope ,
. flowi4_mark = IN_DEV_SRC_VMARK ( in_dev ) ? skb - > mark : 0 ,
} ;
2015-06-23 20:45:37 +03:00
if ( ! fib_lookup ( net , & fl4 , & res , 0 ) )
2012-06-29 05:33:24 +04:00
return FIB_RES_PREFSRC ( net , res ) ;
} else {
scope = RT_SCOPE_LINK ;
}
return inet_select_addr ( dev , ip_hdr ( skb ) - > saddr , scope ) ;
2012-06-28 14:59:11 +04:00
}
2005-04-17 02:20:36 +04:00
/* Given (packet source, input interface) and optional (dst, oif, tos):
2010-10-05 00:00:18 +04:00
* - ( main ) check , that source is valid i . e . not broadcast or our local
* address .
* - figure out what " logical " interface this packet arrived
* and calculate " specific destination " address .
* - check , that packet arrived from expected physical interface .
2010-10-05 14:41:36 +04:00
* called with rcu_read_lock ( )
2005-04-17 02:20:36 +04:00
*/
2012-06-29 12:32:45 +04:00
static int __fib_validate_source ( struct sk_buff * skb , __be32 src , __be32 dst ,
u8 tos , int oif , struct net_device * dev ,
int rpf , struct in_device * idev , u32 * itag )
2005-04-17 02:20:36 +04:00
{
2014-08-17 11:19:54 +04:00
int ret , no_addr ;
2005-04-17 02:20:36 +04:00
struct fib_result res ;
2012-06-29 05:54:02 +04:00
struct flowi4 fl4 ;
2017-02-26 16:50:52 +03:00
struct net * net = dev_net ( dev ) ;
2012-06-29 05:54:02 +04:00
bool dev_match ;
2005-04-17 02:20:36 +04:00
2011-03-12 10:02:42 +03:00
fl4 . flowi4_oif = 0 ;
2015-09-30 06:07:13 +03:00
fl4 . flowi4_iif = l3mdev_master_ifindex_rcu ( dev ) ;
2015-08-13 23:59:01 +03:00
if ( ! fl4 . flowi4_iif )
fl4 . flowi4_iif = oif ? : LOOPBACK_IFINDEX ;
2011-03-12 10:02:42 +03:00
fl4 . daddr = src ;
fl4 . saddr = dst ;
fl4 . flowi4_tos = tos ;
fl4 . flowi4_scope = RT_SCOPE_UNIVERSE ;
2015-07-21 11:43:59 +03:00
fl4 . flowi4_tun_key . tun_id = 0 ;
2015-09-30 05:07:07 +03:00
fl4 . flowi4_flags = 0 ;
2017-02-26 16:50:52 +03:00
fl4 . flowi4_uid = sock_net_uid ( net , NULL ) ;
2011-03-10 07:57:50 +03:00
2012-06-29 05:54:02 +04:00
no_addr = idev - > ifa_list = = NULL ;
2011-04-07 08:51:51 +04:00
2012-06-29 05:54:02 +04:00
fl4 . flowi4_mark = IN_DEV_SRC_VMARK ( idev ) ? skb - > mark : 0 ;
2005-04-17 02:20:36 +04:00
2015-08-28 18:42:09 +03:00
trace_fib_validate_source ( dev , & fl4 ) ;
2015-06-23 20:45:37 +03:00
if ( fib_lookup ( net , & fl4 , & res , 0 ) )
2005-04-17 02:20:36 +04:00
goto last_resort ;
2014-08-17 11:19:54 +04:00
if ( res . type ! = RTN_UNICAST & &
( res . type ! = RTN_LOCAL | | ! IN_DEV_ACCEPT_LOCAL ( idev ) ) )
goto e_inval ;
2017-02-26 16:50:52 +03:00
if ( ! rpf & & ! fib_num_tclassid_users ( net ) & &
2014-08-17 11:19:54 +04:00
( dev - > ifindex ! = oif | | ! IN_DEV_TX_REDIRECTS ( idev ) ) )
goto last_resort ;
2005-04-17 02:20:36 +04:00
fib_combine_itag ( itag , & res ) ;
2010-09-07 09:36:19 +04:00
dev_match = false ;
2005-04-17 02:20:36 +04:00
# ifdef CONFIG_IP_ROUTE_MULTIPATH
2010-09-07 09:36:19 +04:00
for ( ret = 0 ; ret < res . fi - > fib_nhs ; ret + + ) {
struct fib_nh * nh = & res . fi - > fib_nh [ ret ] ;
if ( nh - > nh_dev = = dev ) {
dev_match = true ;
break ;
2015-09-30 06:07:13 +03:00
} else if ( l3mdev_master_ifindex_rcu ( nh - > nh_dev ) = = dev - > ifindex ) {
2015-08-13 23:59:01 +03:00
dev_match = true ;
break ;
2010-09-07 09:36:19 +04:00
}
}
2005-04-17 02:20:36 +04:00
# else
if ( FIB_RES_DEV ( res ) = = dev )
2010-09-07 09:36:19 +04:00
dev_match = true ;
2005-04-17 02:20:36 +04:00
# endif
2010-09-07 09:36:19 +04:00
if ( dev_match ) {
2005-04-17 02:20:36 +04:00
ret = FIB_RES_NH ( res ) . nh_scope > = RT_SCOPE_HOST ;
return ret ;
}
if ( no_addr )
goto last_resort ;
2009-02-20 11:25:36 +03:00
if ( rpf = = 1 )
2010-06-02 16:05:27 +04:00
goto e_rpf ;
2011-03-12 10:02:42 +03:00
fl4 . flowi4_oif = dev - > ifindex ;
2005-04-17 02:20:36 +04:00
ret = 0 ;
2015-06-23 20:45:37 +03:00
if ( fib_lookup ( net , & fl4 , & res , FIB_LOOKUP_IGNORE_LINKSTATE ) = = 0 ) {
2012-06-28 15:05:27 +04:00
if ( res . type = = RTN_UNICAST )
2005-04-17 02:20:36 +04:00
ret = FIB_RES_NH ( res ) . nh_scope > = RT_SCOPE_HOST ;
}
return ret ;
last_resort :
if ( rpf )
2010-06-02 16:05:27 +04:00
goto e_rpf ;
2005-04-17 02:20:36 +04:00
* itag = 0 ;
return 0 ;
e_inval :
return - EINVAL ;
2010-06-02 16:05:27 +04:00
e_rpf :
return - EXDEV ;
2005-04-17 02:20:36 +04:00
}
2012-06-29 12:32:45 +04:00
/* Ignore rp_filter for packets protected by IPsec. */
int fib_validate_source ( struct sk_buff * skb , __be32 src , __be32 dst ,
u8 tos , int oif , struct net_device * dev ,
struct in_device * idev , u32 * itag )
{
int r = secpath_exists ( skb ) ? 0 : IN_DEV_RPFILTER ( idev ) ;
2012-10-08 15:41:15 +04:00
if ( ! r & & ! fib_num_tclassid_users ( dev_net ( dev ) ) & &
2014-08-17 11:19:54 +04:00
IN_DEV_ACCEPT_LOCAL ( idev ) & &
2012-10-08 15:41:15 +04:00
( dev - > ifindex ! = oif | | ! IN_DEV_TX_REDIRECTS ( idev ) ) ) {
2012-06-29 12:32:45 +04:00
* itag = 0 ;
return 0 ;
}
return __fib_validate_source ( skb , src , dst , tos , oif , dev , r , idev , itag ) ;
}
2006-09-28 05:40:00 +04:00
static inline __be32 sk_extract_addr ( struct sockaddr * addr )
2006-08-18 05:14:52 +04:00
{
return ( ( struct sockaddr_in * ) addr ) - > sin_addr . s_addr ;
}
static int put_rtax ( struct nlattr * mx , int len , int type , u32 value )
{
struct nlattr * nla ;
nla = ( struct nlattr * ) ( ( char * ) mx + len ) ;
nla - > nla_type = type ;
nla - > nla_len = nla_attr_size ( 4 ) ;
* ( u32 * ) nla_data ( nla ) = value ;
return len + nla_total_size ( 4 ) ;
}
2008-01-10 14:29:23 +03:00
static int rtentry_to_fib_config ( struct net * net , int cmd , struct rtentry * rt ,
2006-08-18 05:14:52 +04:00
struct fib_config * cfg )
{
2006-09-27 09:15:46 +04:00
__be32 addr ;
2006-08-18 05:14:52 +04:00
int plen ;
memset ( cfg , 0 , sizeof ( * cfg ) ) ;
2008-01-10 14:29:23 +03:00
cfg - > fc_nlinfo . nl_net = net ;
2006-08-18 05:14:52 +04:00
if ( rt - > rt_dst . sa_family ! = AF_INET )
return - EAFNOSUPPORT ;
/*
* Check mask for validity :
* a ) it must be contiguous .
* b ) destination must have all host bits clear .
* c ) if application forgot to set correct family ( AF_INET ) ,
* reject request unless it is absolutely clear i . e .
* both family and mask are zero .
*/
plen = 32 ;
addr = sk_extract_addr ( & rt - > rt_dst ) ;
if ( ! ( rt - > rt_flags & RTF_HOST ) ) {
2006-09-28 05:40:00 +04:00
__be32 mask = sk_extract_addr ( & rt - > rt_genmask ) ;
2006-08-18 05:14:52 +04:00
if ( rt - > rt_genmask . sa_family ! = AF_INET ) {
if ( mask | | rt - > rt_genmask . sa_family )
return - EAFNOSUPPORT ;
}
if ( bad_mask ( mask , addr ) )
return - EINVAL ;
plen = inet_mask_len ( mask ) ;
}
cfg - > fc_dst_len = plen ;
cfg - > fc_dst = addr ;
if ( cmd ! = SIOCDELRT ) {
cfg - > fc_nlflags = NLM_F_CREATE ;
cfg - > fc_protocol = RTPROT_BOOT ;
}
if ( rt - > rt_metric )
cfg - > fc_priority = rt - > rt_metric - 1 ;
if ( rt - > rt_flags & RTF_REJECT ) {
cfg - > fc_scope = RT_SCOPE_HOST ;
cfg - > fc_type = RTN_UNREACHABLE ;
return 0 ;
}
cfg - > fc_scope = RT_SCOPE_NOWHERE ;
cfg - > fc_type = RTN_UNICAST ;
if ( rt - > rt_dev ) {
char * colon ;
struct net_device * dev ;
char devname [ IFNAMSIZ ] ;
if ( copy_from_user ( devname , rt - > rt_dev , IFNAMSIZ - 1 ) )
return - EFAULT ;
devname [ IFNAMSIZ - 1 ] = 0 ;
colon = strchr ( devname , ' : ' ) ;
if ( colon )
* colon = 0 ;
2008-01-10 14:29:23 +03:00
dev = __dev_get_by_name ( net , devname ) ;
2006-08-18 05:14:52 +04:00
if ( ! dev )
return - ENODEV ;
cfg - > fc_oif = dev - > ifindex ;
2016-09-05 01:20:20 +03:00
cfg - > fc_table = l3mdev_fib_table ( dev ) ;
2006-08-18 05:14:52 +04:00
if ( colon ) {
struct in_ifaddr * ifa ;
struct in_device * in_dev = __in_dev_get_rtnl ( dev ) ;
if ( ! in_dev )
return - ENODEV ;
* colon = ' : ' ;
for ( ifa = in_dev - > ifa_list ; ifa ; ifa = ifa - > ifa_next )
if ( strcmp ( ifa - > ifa_label , devname ) = = 0 )
break ;
2015-04-03 11:17:26 +03:00
if ( ! ifa )
2006-08-18 05:14:52 +04:00
return - ENODEV ;
cfg - > fc_prefsrc = ifa - > ifa_local ;
}
}
addr = sk_extract_addr ( & rt - > rt_gateway ) ;
if ( rt - > rt_gateway . sa_family = = AF_INET & & addr ) {
2015-08-13 23:59:05 +03:00
unsigned int addr_type ;
2006-08-18 05:14:52 +04:00
cfg - > fc_gw = addr ;
2015-08-13 23:59:05 +03:00
addr_type = inet_addr_type_table ( net , addr , cfg - > fc_table ) ;
2006-08-18 05:14:52 +04:00
if ( rt - > rt_flags & RTF_GATEWAY & &
2015-08-13 23:59:05 +03:00
addr_type = = RTN_UNICAST )
2006-08-18 05:14:52 +04:00
cfg - > fc_scope = RT_SCOPE_UNIVERSE ;
}
if ( cmd = = SIOCDELRT )
return 0 ;
if ( rt - > rt_flags & RTF_GATEWAY & & ! cfg - > fc_gw )
return - EINVAL ;
if ( cfg - > fc_scope = = RT_SCOPE_NOWHERE )
cfg - > fc_scope = RT_SCOPE_LINK ;
if ( rt - > rt_flags & ( RTF_MTU | RTF_WINDOW | RTF_IRTT ) ) {
struct nlattr * mx ;
int len = 0 ;
mx = kzalloc ( 3 * nla_total_size ( 4 ) , GFP_KERNEL ) ;
2015-04-03 11:17:26 +03:00
if ( ! mx )
2006-08-18 05:14:52 +04:00
return - ENOMEM ;
if ( rt - > rt_flags & RTF_MTU )
len = put_rtax ( mx , len , RTAX_ADVMSS , rt - > rt_mtu - 40 ) ;
if ( rt - > rt_flags & RTF_WINDOW )
len = put_rtax ( mx , len , RTAX_WINDOW , rt - > rt_window ) ;
if ( rt - > rt_flags & RTF_IRTT )
len = put_rtax ( mx , len , RTAX_RTT , rt - > rt_irtt < < 3 ) ;
cfg - > fc_mx = mx ;
cfg - > fc_mx_len = len ;
}
return 0 ;
}
2005-04-17 02:20:36 +04:00
/*
2010-10-05 00:00:18 +04:00
* Handle IP routing ioctl calls .
* These are used to manipulate the routing tables
2005-04-17 02:20:36 +04:00
*/
2008-01-10 14:29:53 +03:00
int ip_rt_ioctl ( struct net * net , unsigned int cmd , void __user * arg )
2005-04-17 02:20:36 +04:00
{
2006-08-18 05:14:52 +04:00
struct fib_config cfg ;
struct rtentry rt ;
2005-04-17 02:20:36 +04:00
int err ;
switch ( cmd ) {
case SIOCADDRT : /* Add a route */
case SIOCDELRT : /* Delete a route */
net: Allow userns root to control ipv4
Allow an unpriviled user who has created a user namespace, and then
created a network namespace to effectively use the new network
namespace, by reducing capable(CAP_NET_ADMIN) and
capable(CAP_NET_RAW) calls to be ns_capable(net->user_ns,
CAP_NET_ADMIN), or capable(net->user_ns, CAP_NET_RAW) calls.
Settings that merely control a single network device are allowed.
Either the network device is a logical network device where
restrictions make no difference or the network device is hardware NIC
that has been explicity moved from the initial network namespace.
In general policy and network stack state changes are allowed
while resource control is left unchanged.
Allow creating raw sockets.
Allow the SIOCSARP ioctl to control the arp cache.
Allow the SIOCSIFFLAG ioctl to allow setting network device flags.
Allow the SIOCSIFADDR ioctl to allow setting a netdevice ipv4 address.
Allow the SIOCSIFBRDADDR ioctl to allow setting a netdevice ipv4 broadcast address.
Allow the SIOCSIFDSTADDR ioctl to allow setting a netdevice ipv4 destination address.
Allow the SIOCSIFNETMASK ioctl to allow setting a netdevice ipv4 netmask.
Allow the SIOCADDRT and SIOCDELRT ioctls to allow adding and deleting ipv4 routes.
Allow the SIOCADDTUNNEL, SIOCCHGTUNNEL and SIOCDELTUNNEL ioctls for
adding, changing and deleting gre tunnels.
Allow the SIOCADDTUNNEL, SIOCCHGTUNNEL and SIOCDELTUNNEL ioctls for
adding, changing and deleting ipip tunnels.
Allow the SIOCADDTUNNEL, SIOCCHGTUNNEL and SIOCDELTUNNEL ioctls for
adding, changing and deleting ipsec virtual tunnel interfaces.
Allow setting the MRT_INIT, MRT_DONE, MRT_ADD_VIF, MRT_DEL_VIF, MRT_ADD_MFC,
MRT_DEL_MFC, MRT_ASSERT, MRT_PIM, MRT_TABLE socket options on multicast routing
sockets.
Allow setting and receiving IPOPT_CIPSO, IP_OPT_SEC, IP_OPT_SID and
arbitrary ip options.
Allow setting IP_SEC_POLICY/IP_XFRM_POLICY ipv4 socket option.
Allow setting the IP_TRANSPARENT ipv4 socket option.
Allow setting the TCP_REPAIR socket option.
Allow setting the TCP_CONGESTION socket option.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-16 07:03:05 +04:00
if ( ! ns_capable ( net - > user_ns , CAP_NET_ADMIN ) )
2005-04-17 02:20:36 +04:00
return - EPERM ;
2006-08-18 05:14:52 +04:00
if ( copy_from_user ( & rt , arg , sizeof ( rt ) ) )
2005-04-17 02:20:36 +04:00
return - EFAULT ;
2006-08-18 05:14:52 +04:00
2005-04-17 02:20:36 +04:00
rtnl_lock ( ) ;
2008-01-10 14:29:53 +03:00
err = rtentry_to_fib_config ( net , cmd , & rt , & cfg ) ;
2005-04-17 02:20:36 +04:00
if ( err = = 0 ) {
2006-08-18 05:14:52 +04:00
struct fib_table * tb ;
2005-04-17 02:20:36 +04:00
if ( cmd = = SIOCDELRT ) {
2008-01-10 14:29:53 +03:00
tb = fib_get_table ( net , cfg . fc_table ) ;
2005-04-17 02:20:36 +04:00
if ( tb )
2017-05-28 01:19:26 +03:00
err = fib_table_delete ( net , tb , & cfg ,
NULL ) ;
2006-08-18 05:14:52 +04:00
else
err = - ESRCH ;
2005-04-17 02:20:36 +04:00
} else {
2008-01-10 14:29:53 +03:00
tb = fib_new_table ( net , cfg . fc_table ) ;
2005-04-17 02:20:36 +04:00
if ( tb )
2017-05-21 19:12:02 +03:00
err = fib_table_insert ( net , tb ,
& cfg , NULL ) ;
2006-08-18 05:14:52 +04:00
else
err = - ENOBUFS ;
2005-04-17 02:20:36 +04:00
}
2006-08-18 05:14:52 +04:00
/* allocated by rtentry_to_fib_config() */
kfree ( cfg . fc_mx ) ;
2005-04-17 02:20:36 +04:00
}
rtnl_unlock ( ) ;
return err ;
}
return - EINVAL ;
}
2010-10-05 00:00:18 +04:00
const struct nla_policy rtm_ipv4_policy [ RTA_MAX + 1 ] = {
2006-08-18 05:14:52 +04:00
[ RTA_DST ] = { . type = NLA_U32 } ,
[ RTA_SRC ] = { . type = NLA_U32 } ,
[ RTA_IIF ] = { . type = NLA_U32 } ,
[ RTA_OIF ] = { . type = NLA_U32 } ,
[ RTA_GATEWAY ] = { . type = NLA_U32 } ,
[ RTA_PRIORITY ] = { . type = NLA_U32 } ,
[ RTA_PREFSRC ] = { . type = NLA_U32 } ,
[ RTA_METRICS ] = { . type = NLA_NESTED } ,
2006-08-27 07:13:18 +04:00
[ RTA_MULTIPATH ] = { . len = sizeof ( struct rtnexthop ) } ,
2006-08-18 05:14:52 +04:00
[ RTA_FLOW ] = { . type = NLA_U32 } ,
2015-07-21 11:43:47 +03:00
[ RTA_ENCAP_TYPE ] = { . type = NLA_U16 } ,
[ RTA_ENCAP ] = { . type = NLA_NESTED } ,
2016-11-03 20:23:42 +03:00
[ RTA_UID ] = { . type = NLA_U32 } ,
2017-02-27 15:59:39 +03:00
[ RTA_MARK ] = { . type = NLA_U32 } ,
2006-08-18 05:14:52 +04:00
} ;
2008-01-10 14:29:23 +03:00
static int rtm_to_fib_config ( struct net * net , struct sk_buff * skb ,
2017-05-21 19:12:02 +03:00
struct nlmsghdr * nlh , struct fib_config * cfg ,
struct netlink_ext_ack * extack )
2005-04-17 02:20:36 +04:00
{
2006-08-18 05:14:52 +04:00
struct nlattr * attr ;
int err , remaining ;
struct rtmsg * rtm ;
2017-04-12 15:34:07 +03:00
err = nlmsg_validate ( nlh , sizeof ( * rtm ) , RTA_MAX , rtm_ipv4_policy ,
2017-05-21 19:12:02 +03:00
extack ) ;
2006-08-18 05:14:52 +04:00
if ( err < 0 )
goto errout ;
memset ( cfg , 0 , sizeof ( * cfg ) ) ;
rtm = nlmsg_data ( nlh ) ;
cfg - > fc_dst_len = rtm - > rtm_dst_len ;
cfg - > fc_tos = rtm - > rtm_tos ;
cfg - > fc_table = rtm - > rtm_table ;
cfg - > fc_protocol = rtm - > rtm_protocol ;
cfg - > fc_scope = rtm - > rtm_scope ;
cfg - > fc_type = rtm - > rtm_type ;
cfg - > fc_flags = rtm - > rtm_flags ;
cfg - > fc_nlflags = nlh - > nlmsg_flags ;
2012-09-08 00:12:54 +04:00
cfg - > fc_nlinfo . portid = NETLINK_CB ( skb ) . portid ;
2006-08-18 05:14:52 +04:00
cfg - > fc_nlinfo . nlh = nlh ;
2008-01-10 14:29:23 +03:00
cfg - > fc_nlinfo . nl_net = net ;
2006-08-18 05:14:52 +04:00
2007-03-25 07:32:54 +04:00
if ( cfg - > fc_type > RTN_MAX ) {
2017-05-21 19:12:03 +03:00
NL_SET_ERR_MSG ( extack , " Invalid route type " ) ;
2007-03-25 07:32:54 +04:00
err = - EINVAL ;
goto errout ;
}
2006-08-18 05:14:52 +04:00
nlmsg_for_each_attr ( attr , nlh , sizeof ( struct rtmsg ) , remaining ) {
2007-09-12 16:44:36 +04:00
switch ( nla_type ( attr ) ) {
2006-08-18 05:14:52 +04:00
case RTA_DST :
2006-09-27 09:15:25 +04:00
cfg - > fc_dst = nla_get_be32 ( attr ) ;
2006-08-18 05:14:52 +04:00
break ;
case RTA_OIF :
cfg - > fc_oif = nla_get_u32 ( attr ) ;
break ;
case RTA_GATEWAY :
2006-09-27 09:15:25 +04:00
cfg - > fc_gw = nla_get_be32 ( attr ) ;
2006-08-18 05:14:52 +04:00
break ;
case RTA_PRIORITY :
cfg - > fc_priority = nla_get_u32 ( attr ) ;
break ;
case RTA_PREFSRC :
2006-09-27 09:15:25 +04:00
cfg - > fc_prefsrc = nla_get_be32 ( attr ) ;
2006-08-18 05:14:52 +04:00
break ;
case RTA_METRICS :
cfg - > fc_mx = nla_data ( attr ) ;
cfg - > fc_mx_len = nla_len ( attr ) ;
break ;
case RTA_MULTIPATH :
2017-01-18 01:57:36 +03:00
err = lwtunnel_valid_encap_type_attr ( nla_data ( attr ) ,
nla_len ( attr ) ) ;
if ( err < 0 )
goto errout ;
2006-08-18 05:14:52 +04:00
cfg - > fc_mp = nla_data ( attr ) ;
cfg - > fc_mp_len = nla_len ( attr ) ;
break ;
case RTA_FLOW :
cfg - > fc_flow = nla_get_u32 ( attr ) ;
break ;
case RTA_TABLE :
cfg - > fc_table = nla_get_u32 ( attr ) ;
break ;
2015-07-21 11:43:47 +03:00
case RTA_ENCAP :
cfg - > fc_encap = attr ;
break ;
case RTA_ENCAP_TYPE :
cfg - > fc_encap_type = nla_get_u16 ( attr ) ;
2017-01-18 01:57:36 +03:00
err = lwtunnel_valid_encap_type ( cfg - > fc_encap_type ) ;
if ( err < 0 )
goto errout ;
2015-07-21 11:43:47 +03:00
break ;
2005-04-17 02:20:36 +04:00
}
}
2006-08-18 05:14:52 +04:00
2005-04-17 02:20:36 +04:00
return 0 ;
2006-08-18 05:14:52 +04:00
errout :
return err ;
2005-04-17 02:20:36 +04:00
}
2017-04-16 19:48:24 +03:00
static int inet_rtm_delroute ( struct sk_buff * skb , struct nlmsghdr * nlh ,
struct netlink_ext_ack * extack )
2005-04-17 02:20:36 +04:00
{
2008-03-25 20:26:21 +03:00
struct net * net = sock_net ( skb - > sk ) ;
2006-08-18 05:14:52 +04:00
struct fib_config cfg ;
struct fib_table * tb ;
int err ;
2005-04-17 02:20:36 +04:00
2017-05-21 19:12:02 +03:00
err = rtm_to_fib_config ( net , skb , nlh , & cfg , extack ) ;
2006-08-18 05:14:52 +04:00
if ( err < 0 )
goto errout ;
2005-04-17 02:20:36 +04:00
2008-01-10 14:24:11 +03:00
tb = fib_get_table ( net , cfg . fc_table ) ;
2015-04-03 11:17:26 +03:00
if ( ! tb ) {
2017-05-21 19:12:03 +03:00
NL_SET_ERR_MSG ( extack , " FIB table does not exist " ) ;
2006-08-18 05:14:52 +04:00
err = - ESRCH ;
goto errout ;
}
2017-05-28 01:19:26 +03:00
err = fib_table_delete ( net , tb , & cfg , extack ) ;
2006-08-18 05:14:52 +04:00
errout :
return err ;
2005-04-17 02:20:36 +04:00
}
2017-04-16 19:48:24 +03:00
static int inet_rtm_newroute ( struct sk_buff * skb , struct nlmsghdr * nlh ,
struct netlink_ext_ack * extack )
2005-04-17 02:20:36 +04:00
{
2008-03-25 20:26:21 +03:00
struct net * net = sock_net ( skb - > sk ) ;
2006-08-18 05:14:52 +04:00
struct fib_config cfg ;
struct fib_table * tb ;
int err ;
2005-04-17 02:20:36 +04:00
2017-05-21 19:12:02 +03:00
err = rtm_to_fib_config ( net , skb , nlh , & cfg , extack ) ;
2006-08-18 05:14:52 +04:00
if ( err < 0 )
goto errout ;
2005-04-17 02:20:36 +04:00
2008-01-10 14:30:24 +03:00
tb = fib_new_table ( net , cfg . fc_table ) ;
2015-04-03 11:17:26 +03:00
if ( ! tb ) {
2006-08-18 05:14:52 +04:00
err = - ENOBUFS ;
goto errout ;
}
2017-05-21 19:12:02 +03:00
err = fib_table_insert ( net , tb , & cfg , extack ) ;
2006-08-18 05:14:52 +04:00
errout :
return err ;
2005-04-17 02:20:36 +04:00
}
2007-03-22 21:55:17 +03:00
static int inet_dump_fib ( struct sk_buff * skb , struct netlink_callback * cb )
2005-04-17 02:20:36 +04:00
{
2008-03-25 20:26:21 +03:00
struct net * net = sock_net ( skb - > sk ) ;
2006-08-11 10:10:46 +04:00
unsigned int h , s_h ;
unsigned int e = 0 , s_e ;
2005-04-17 02:20:36 +04:00
struct fib_table * tb ;
2008-01-10 14:28:24 +03:00
struct hlist_head * head ;
2017-05-16 09:19:17 +03:00
int dumped = 0 , err ;
2005-04-17 02:20:36 +04:00
2006-08-18 05:15:17 +04:00
if ( nlmsg_len ( cb - > nlh ) > = sizeof ( struct rtmsg ) & &
( ( struct rtmsg * ) nlmsg_data ( cb - > nlh ) ) - > rtm_flags & RTM_F_CLONED )
2014-03-21 07:33:10 +04:00
return skb - > len ;
2005-04-17 02:20:36 +04:00
2006-08-11 10:10:46 +04:00
s_h = cb - > args [ 0 ] ;
s_e = cb - > args [ 1 ] ;
2015-03-05 02:02:44 +03:00
rcu_read_lock ( ) ;
2006-08-11 10:10:46 +04:00
for ( h = s_h ; h < FIB_TABLE_HASHSZ ; h + + , s_e = 0 ) {
e = 0 ;
2008-01-10 14:28:24 +03:00
head = & net - > ipv4 . fib_table_hash [ h ] ;
2015-03-05 02:02:44 +03:00
hlist_for_each_entry_rcu ( tb , head , tb_hlist ) {
2006-08-11 10:10:46 +04:00
if ( e < s_e )
goto next ;
if ( dumped )
memset ( & cb - > args [ 2 ] , 0 , sizeof ( cb - > args ) -
2007-02-09 17:24:47 +03:00
2 * sizeof ( cb - > args [ 0 ] ) ) ;
2017-05-16 09:19:17 +03:00
err = fib_table_dump ( tb , skb , cb ) ;
if ( err < 0 ) {
if ( likely ( skb - > len ) )
goto out ;
goto out_err ;
}
2006-08-11 10:10:46 +04:00
dumped = 1 ;
next :
e + + ;
}
2005-04-17 02:20:36 +04:00
}
2006-08-11 10:10:46 +04:00
out :
2017-05-16 09:19:17 +03:00
err = skb - > len ;
out_err :
2015-03-05 02:02:44 +03:00
rcu_read_unlock ( ) ;
2006-08-11 10:10:46 +04:00
cb - > args [ 1 ] = e ;
cb - > args [ 0 ] = h ;
2005-04-17 02:20:36 +04:00
2017-05-16 09:19:17 +03:00
return err ;
2005-04-17 02:20:36 +04:00
}
/* Prepare and feed intra-kernel routing request.
2010-10-05 00:00:18 +04:00
* Really , it should be netlink message , but : - ( netlink
* can be not configured , so that we feed it directly
* to fib engine . It is legal , because all events occur
* only when netlink is already locked .
2005-04-17 02:20:36 +04:00
*/
2006-09-28 05:40:00 +04:00
static void fib_magic ( int cmd , int type , __be32 dst , int dst_len , struct in_ifaddr * ifa )
2005-04-17 02:20:36 +04:00
{
2008-03-25 15:47:49 +03:00
struct net * net = dev_net ( ifa - > ifa_dev - > dev ) ;
2015-09-30 06:07:14 +03:00
u32 tb_id = l3mdev_fib_table ( ifa - > ifa_dev - > dev ) ;
2006-08-18 05:14:52 +04:00
struct fib_table * tb ;
struct fib_config cfg = {
. fc_protocol = RTPROT_KERNEL ,
. fc_type = type ,
. fc_dst = dst ,
. fc_dst_len = dst_len ,
. fc_prefsrc = ifa - > ifa_local ,
. fc_oif = ifa - > ifa_dev - > dev - > ifindex ,
. fc_nlflags = NLM_F_CREATE | NLM_F_APPEND ,
2008-01-10 14:26:13 +03:00
. fc_nlinfo = {
2008-01-10 14:29:23 +03:00
. nl_net = net ,
2008-01-10 14:26:13 +03:00
} ,
2006-08-18 05:14:52 +04:00
} ;
2005-04-17 02:20:36 +04:00
2015-08-13 23:59:06 +03:00
if ( ! tb_id )
tb_id = ( type = = RTN_UNICAST ) ? RT_TABLE_MAIN : RT_TABLE_LOCAL ;
2005-04-17 02:20:36 +04:00
2015-08-13 23:59:06 +03:00
tb = fib_new_table ( net , tb_id ) ;
2015-04-03 11:17:26 +03:00
if ( ! tb )
2005-04-17 02:20:36 +04:00
return ;
2006-08-18 05:14:52 +04:00
cfg . fc_table = tb - > tb_id ;
2005-04-17 02:20:36 +04:00
2006-08-18 05:14:52 +04:00
if ( type ! = RTN_LOCAL )
cfg . fc_scope = RT_SCOPE_LINK ;
else
cfg . fc_scope = RT_SCOPE_HOST ;
2005-04-17 02:20:36 +04:00
if ( cmd = = RTM_NEWROUTE )
2017-05-21 19:12:02 +03:00
fib_table_insert ( net , tb , & cfg , NULL ) ;
2005-04-17 02:20:36 +04:00
else
2017-05-28 01:19:26 +03:00
fib_table_delete ( net , tb , & cfg , NULL ) ;
2005-04-17 02:20:36 +04:00
}
2005-11-23 01:47:37 +03:00
void fib_add_ifaddr ( struct in_ifaddr * ifa )
2005-04-17 02:20:36 +04:00
{
struct in_device * in_dev = ifa - > ifa_dev ;
struct net_device * dev = in_dev - > dev ;
struct in_ifaddr * prim = ifa ;
2006-09-29 05:00:55 +04:00
__be32 mask = ifa - > ifa_mask ;
__be32 addr = ifa - > ifa_local ;
2010-10-05 00:00:18 +04:00
__be32 prefix = ifa - > ifa_address & mask ;
2005-04-17 02:20:36 +04:00
2010-10-05 00:00:18 +04:00
if ( ifa - > ifa_flags & IFA_F_SECONDARY ) {
2005-04-17 02:20:36 +04:00
prim = inet_ifa_byprefix ( in_dev , prefix , mask ) ;
2015-04-03 11:17:26 +03:00
if ( ! prim ) {
2012-03-11 22:36:11 +04:00
pr_warn ( " %s: bug: prim == NULL \n " , __func__ ) ;
2005-04-17 02:20:36 +04:00
return ;
}
}
fib_magic ( RTM_NEWROUTE , RTN_LOCAL , addr , 32 , prim ) ;
2010-10-05 00:00:18 +04:00
if ( ! ( dev - > flags & IFF_UP ) )
2005-04-17 02:20:36 +04:00
return ;
/* Add broadcast address, if it is explicitly assigned. */
2006-09-29 05:00:55 +04:00
if ( ifa - > ifa_broadcast & & ifa - > ifa_broadcast ! = htonl ( 0xFFFFFFFF ) )
2005-04-17 02:20:36 +04:00
fib_magic ( RTM_NEWROUTE , RTN_BROADCAST , ifa - > ifa_broadcast , 32 , prim ) ;
2010-10-05 00:00:18 +04:00
if ( ! ipv4_is_zeronet ( prefix ) & & ! ( ifa - > ifa_flags & IFA_F_SECONDARY ) & &
2005-04-17 02:20:36 +04:00
( prefix ! = addr | | ifa - > ifa_prefixlen < 32 ) ) {
2015-10-20 11:28:45 +03:00
if ( ! ( ifa - > ifa_flags & IFA_F_NOPREFIXROUTE ) )
fib_magic ( RTM_NEWROUTE ,
dev - > flags & IFF_LOOPBACK ? RTN_LOCAL : RTN_UNICAST ,
prefix , ifa - > ifa_prefixlen , prim ) ;
2005-04-17 02:20:36 +04:00
/* Add network specific broadcasts, when it takes a sense */
if ( ifa - > ifa_prefixlen < 31 ) {
fib_magic ( RTM_NEWROUTE , RTN_BROADCAST , prefix , 32 , prim ) ;
2010-10-05 00:00:18 +04:00
fib_magic ( RTM_NEWROUTE , RTN_BROADCAST , prefix | ~ mask ,
32 , prim ) ;
2005-04-17 02:20:36 +04:00
}
}
}
2011-03-19 15:13:49 +03:00
/* Delete primary or secondary address.
* Optionally , on secondary address promotion consider the addresses
* from subnet iprim as deleted , even if they are in device list .
* In this case the secondary ifa can be in device list .
*/
void fib_del_ifaddr ( struct in_ifaddr * ifa , struct in_ifaddr * iprim )
2005-04-17 02:20:36 +04:00
{
struct in_device * in_dev = ifa - > ifa_dev ;
struct net_device * dev = in_dev - > dev ;
struct in_ifaddr * ifa1 ;
2011-03-19 15:13:49 +03:00
struct in_ifaddr * prim = ifa , * prim1 = NULL ;
2010-10-05 00:00:18 +04:00
__be32 brd = ifa - > ifa_address | ~ ifa - > ifa_mask ;
__be32 any = ifa - > ifa_address & ifa - > ifa_mask ;
2005-04-17 02:20:36 +04:00
# define LOCAL_OK 1
# define BRD_OK 2
# define BRD0_OK 4
# define BRD1_OK 8
2012-04-15 09:58:06 +04:00
unsigned int ok = 0 ;
2011-03-19 15:13:49 +03:00
int subnet = 0 ; /* Primary network */
int gone = 1 ; /* Address is missing */
int same_prefsrc = 0 ; /* Another primary with same IP */
2005-04-17 02:20:36 +04:00
2011-03-19 15:13:49 +03:00
if ( ifa - > ifa_flags & IFA_F_SECONDARY ) {
2005-04-17 02:20:36 +04:00
prim = inet_ifa_byprefix ( in_dev , any , ifa - > ifa_mask ) ;
2015-04-03 11:17:26 +03:00
if ( ! prim ) {
2016-04-21 23:23:31 +03:00
/* if the device has been deleted, we don't perform
* address promotion
*/
if ( ! in_dev - > dead )
pr_warn ( " %s: bug: prim == NULL \n " , __func__ ) ;
2005-04-17 02:20:36 +04:00
return ;
}
2011-03-19 15:13:49 +03:00
if ( iprim & & iprim ! = prim ) {
2012-03-11 22:36:11 +04:00
pr_warn ( " %s: bug: iprim != prim \n " , __func__ ) ;
2011-03-19 15:13:49 +03:00
return ;
}
} else if ( ! ipv4_is_zeronet ( any ) & &
( any ! = ifa - > ifa_local | | ifa - > ifa_prefixlen < 32 ) ) {
2015-10-20 11:28:45 +03:00
if ( ! ( ifa - > ifa_flags & IFA_F_NOPREFIXROUTE ) )
fib_magic ( RTM_DELROUTE ,
dev - > flags & IFF_LOOPBACK ? RTN_LOCAL : RTN_UNICAST ,
any , ifa - > ifa_prefixlen , prim ) ;
2011-03-19 15:13:49 +03:00
subnet = 1 ;
2005-04-17 02:20:36 +04:00
}
2016-03-14 06:28:00 +03:00
if ( in_dev - > dead )
goto no_promotions ;
2005-04-17 02:20:36 +04:00
/* Deletion is more complicated than add.
2010-10-05 00:00:18 +04:00
* We should take care of not to delete too much : - )
*
* Scan address list to be sure that addresses are really gone .
2005-04-17 02:20:36 +04:00
*/
for ( ifa1 = in_dev - > ifa_list ; ifa1 ; ifa1 = ifa1 - > ifa_next ) {
2011-03-19 15:13:49 +03:00
if ( ifa1 = = ifa ) {
/* promotion, keep the IP */
gone = 0 ;
continue ;
}
/* Ignore IFAs from our subnet */
if ( iprim & & ifa1 - > ifa_mask = = iprim - > ifa_mask & &
inet_ifa_match ( ifa1 - > ifa_address , iprim ) )
continue ;
/* Ignore ifa1 if it uses different primary IP (prefsrc) */
if ( ifa1 - > ifa_flags & IFA_F_SECONDARY ) {
/* Another address from our subnet? */
if ( ifa1 - > ifa_mask = = prim - > ifa_mask & &
inet_ifa_match ( ifa1 - > ifa_address , prim ) )
prim1 = prim ;
else {
/* We reached the secondaries, so
* same_prefsrc should be determined .
*/
if ( ! same_prefsrc )
continue ;
/* Search new prim1 if ifa1 is not
* using the current prim1
*/
if ( ! prim1 | |
ifa1 - > ifa_mask ! = prim1 - > ifa_mask | |
! inet_ifa_match ( ifa1 - > ifa_address , prim1 ) )
prim1 = inet_ifa_byprefix ( in_dev ,
ifa1 - > ifa_address ,
ifa1 - > ifa_mask ) ;
if ( ! prim1 )
continue ;
if ( prim1 - > ifa_local ! = prim - > ifa_local )
continue ;
}
} else {
if ( prim - > ifa_local ! = ifa1 - > ifa_local )
continue ;
prim1 = ifa1 ;
if ( prim ! = prim1 )
same_prefsrc = 1 ;
}
2005-04-17 02:20:36 +04:00
if ( ifa - > ifa_local = = ifa1 - > ifa_local )
ok | = LOCAL_OK ;
if ( ifa - > ifa_broadcast = = ifa1 - > ifa_broadcast )
ok | = BRD_OK ;
if ( brd = = ifa1 - > ifa_broadcast )
ok | = BRD1_OK ;
if ( any = = ifa1 - > ifa_broadcast )
ok | = BRD0_OK ;
2011-03-19 15:13:49 +03:00
/* primary has network specific broadcasts */
if ( prim1 = = ifa1 & & ifa1 - > ifa_prefixlen < 31 ) {
__be32 brd1 = ifa1 - > ifa_address | ~ ifa1 - > ifa_mask ;
__be32 any1 = ifa1 - > ifa_address & ifa1 - > ifa_mask ;
if ( ! ipv4_is_zeronet ( any1 ) ) {
if ( ifa - > ifa_broadcast = = brd1 | |
ifa - > ifa_broadcast = = any1 )
ok | = BRD_OK ;
if ( brd = = brd1 | | brd = = any1 )
ok | = BRD1_OK ;
if ( any = = brd1 | | any = = any1 )
ok | = BRD0_OK ;
}
}
2005-04-17 02:20:36 +04:00
}
2016-03-14 06:28:00 +03:00
no_promotions :
2010-10-05 00:00:18 +04:00
if ( ! ( ok & BRD_OK ) )
2005-04-17 02:20:36 +04:00
fib_magic ( RTM_DELROUTE , RTN_BROADCAST , ifa - > ifa_broadcast , 32 , prim ) ;
2011-03-19 15:13:49 +03:00
if ( subnet & & ifa - > ifa_prefixlen < 31 ) {
if ( ! ( ok & BRD1_OK ) )
fib_magic ( RTM_DELROUTE , RTN_BROADCAST , brd , 32 , prim ) ;
if ( ! ( ok & BRD0_OK ) )
fib_magic ( RTM_DELROUTE , RTN_BROADCAST , any , 32 , prim ) ;
}
2010-10-05 00:00:18 +04:00
if ( ! ( ok & LOCAL_OK ) ) {
2015-08-13 23:59:05 +03:00
unsigned int addr_type ;
2005-04-17 02:20:36 +04:00
fib_magic ( RTM_DELROUTE , RTN_LOCAL , ifa - > ifa_local , 32 , prim ) ;
/* Check, that this local address finally disappeared. */
2015-08-13 23:59:05 +03:00
addr_type = inet_addr_type_dev_table ( dev_net ( dev ) , dev ,
ifa - > ifa_local ) ;
if ( gone & & addr_type ! = RTN_LOCAL ) {
2005-04-17 02:20:36 +04:00
/* And the last, but not the least thing.
2010-10-05 00:00:18 +04:00
* We must flush stray FIB entries .
*
* First of all , we scan fib_info list searching
* for stray nexthop entries , then ignite fib_flush .
*/
2016-09-05 01:20:20 +03:00
if ( fib_sync_down_addr ( dev , ifa - > ifa_local ) )
2008-03-25 15:47:49 +03:00
fib_flush ( dev_net ( dev ) ) ;
2005-04-17 02:20:36 +04:00
}
}
# undef LOCAL_OK
# undef BRD_OK
# undef BRD0_OK
# undef BRD1_OK
}
2014-12-31 21:56:24 +03:00
static void nl_fib_lookup ( struct net * net , struct fib_result_nl * frn )
2005-06-21 00:36:39 +04:00
{
2007-02-09 17:24:47 +03:00
2005-06-21 00:36:39 +04:00
struct fib_result res ;
2011-03-12 10:02:42 +03:00
struct flowi4 fl4 = {
. flowi4_mark = frn - > fl_mark ,
. daddr = frn - > fl_addr ,
. flowi4_tos = frn - > fl_tos ,
. flowi4_scope = frn - > fl_scope ,
2010-10-05 00:00:18 +04:00
} ;
2014-12-31 21:56:24 +03:00
struct fib_table * tb ;
rcu_read_lock ( ) ;
tb = fib_get_table ( net , frn - > tb_id_in ) ;
2007-04-26 00:07:28 +04:00
frn - > err = - ENOENT ;
2005-06-21 00:36:39 +04:00
if ( tb ) {
local_bh_disable ( ) ;
frn - > tb_id = tb - > tb_id ;
2011-03-12 10:02:42 +03:00
frn - > err = fib_table_lookup ( tb , & fl4 , & res , FIB_LOOKUP_NOREF ) ;
2005-06-21 00:36:39 +04:00
if ( ! frn - > err ) {
frn - > prefixlen = res . prefixlen ;
frn - > nh_sel = res . nh_sel ;
frn - > type = res . type ;
frn - > scope = res . scope ;
}
local_bh_enable ( ) ;
}
2014-12-31 21:56:24 +03:00
rcu_read_unlock ( ) ;
2005-06-21 00:36:39 +04:00
}
2007-10-11 08:32:39 +04:00
static void nl_fib_input ( struct sk_buff * skb )
2005-06-21 00:36:39 +04:00
{
2008-01-10 14:28:55 +03:00
struct net * net ;
2005-06-21 00:36:39 +04:00
struct fib_result_nl * frn ;
2007-10-11 08:32:39 +04:00
struct nlmsghdr * nlh ;
2012-09-08 00:12:54 +04:00
u32 portid ;
2007-04-26 00:07:28 +04:00
2008-03-25 20:26:21 +03:00
net = sock_net ( skb - > sk ) ;
2007-04-26 06:08:35 +04:00
nlh = nlmsg_hdr ( skb ) ;
2017-03-22 05:22:28 +03:00
if ( skb - > len < nlmsg_total_size ( sizeof ( * frn ) ) | |
skb - > len < nlh - > nlmsg_len | |
2013-03-27 10:47:04 +04:00
nlmsg_len ( nlh ) < sizeof ( * frn ) )
2005-12-02 01:30:00 +03:00
return ;
2007-12-21 13:01:53 +03:00
2013-06-28 05:04:23 +04:00
skb = netlink_skb_clone ( skb , GFP_KERNEL ) ;
2015-04-03 11:17:26 +03:00
if ( ! skb )
2007-12-21 13:01:53 +03:00
return ;
nlh = nlmsg_hdr ( skb ) ;
2007-02-09 17:24:47 +03:00
2013-03-27 10:47:04 +04:00
frn = ( struct fib_result_nl * ) nlmsg_data ( nlh ) ;
2014-12-31 21:56:24 +03:00
nl_fib_lookup ( net , frn ) ;
2007-02-09 17:24:47 +03:00
2013-01-11 13:59:20 +04:00
portid = NETLINK_CB ( skb ) . portid ; /* netlink portid */
2012-09-08 00:12:54 +04:00
NETLINK_CB ( skb ) . portid = 0 ; /* from kernel */
2005-08-15 06:29:52 +04:00
NETLINK_CB ( skb ) . dst_group = 0 ; /* unicast */
2012-09-08 00:12:54 +04:00
netlink_unicast ( net - > ipv4 . fibnl , skb , portid , MSG_DONTWAIT ) ;
2007-02-09 17:24:47 +03:00
}
2005-06-21 00:36:39 +04:00
2010-01-17 06:35:32 +03:00
static int __net_init nl_fib_lookup_init ( struct net * net )
2005-06-21 00:36:39 +04:00
{
2008-01-10 14:28:55 +03:00
struct sock * sk ;
2012-06-29 10:15:21 +04:00
struct netlink_kernel_cfg cfg = {
. input = nl_fib_input ,
} ;
2012-09-08 06:53:54 +04:00
sk = netlink_kernel_create ( net , NETLINK_FIB_LOOKUP , & cfg ) ;
2015-04-03 11:17:26 +03:00
if ( ! sk )
2008-01-10 14:22:17 +03:00
return - EAFNOSUPPORT ;
2008-01-10 14:28:55 +03:00
net - > ipv4 . fibnl = sk ;
2008-01-10 14:22:17 +03:00
return 0 ;
}
static void nl_fib_lookup_exit ( struct net * net )
{
2008-01-29 01:41:19 +03:00
netlink_kernel_release ( net - > ipv4 . fibnl ) ;
2008-01-19 10:55:19 +03:00
net - > ipv4 . fibnl = NULL ;
2005-06-21 00:36:39 +04:00
}
2015-10-30 11:23:33 +03:00
static void fib_disable_ip ( struct net_device * dev , unsigned long event ,
bool force )
2005-04-17 02:20:36 +04:00
{
2015-10-30 11:23:33 +03:00
if ( fib_sync_down_dev ( dev , event , force ) )
2008-03-25 15:47:49 +03:00
fib_flush ( dev_net ( dev ) ) ;
2017-04-26 14:04:04 +03:00
else
rt_cache_flush ( dev_net ( dev ) ) ;
2005-04-17 02:20:36 +04:00
arp_ifdown ( dev ) ;
}
static int fib_inetaddr_event ( struct notifier_block * this , unsigned long event , void * ptr )
{
2008-11-03 11:25:16 +03:00
struct in_ifaddr * ifa = ( struct in_ifaddr * ) ptr ;
2008-07-06 06:00:44 +04:00
struct net_device * dev = ifa - > ifa_dev - > dev ;
2011-03-25 03:42:21 +03:00
struct net * net = dev_net ( dev ) ;
2005-04-17 02:20:36 +04:00
switch ( event ) {
case NETDEV_UP :
fib_add_ifaddr ( ifa ) ;
# ifdef CONFIG_IP_ROUTE_MULTIPATH
2015-06-23 20:45:36 +03:00
fib_sync_up ( dev , RTNH_F_DEAD ) ;
2005-04-17 02:20:36 +04:00
# endif
2011-03-25 03:42:21 +03:00
atomic_inc ( & net - > ipv4 . dev_addr_genid ) ;
2012-09-07 04:45:29 +04:00
rt_cache_flush ( dev_net ( dev ) ) ;
2005-04-17 02:20:36 +04:00
break ;
case NETDEV_DOWN :
2011-03-19 15:13:49 +03:00
fib_del_ifaddr ( ifa , NULL ) ;
2011-03-25 03:42:21 +03:00
atomic_inc ( & net - > ipv4 . dev_addr_genid ) ;
2015-04-03 11:17:26 +03:00
if ( ! ifa - > ifa_dev - > ifa_list ) {
2005-04-17 02:20:36 +04:00
/* Last address was deleted from this interface.
2010-10-05 00:00:18 +04:00
* Disable IP .
2005-04-17 02:20:36 +04:00
*/
2015-10-30 11:23:33 +03:00
fib_disable_ip ( dev , event , true ) ;
2005-04-17 02:20:36 +04:00
} else {
2012-09-07 04:45:29 +04:00
rt_cache_flush ( dev_net ( dev ) ) ;
2005-04-17 02:20:36 +04:00
}
break ;
}
return NOTIFY_DONE ;
}
static int fib_netdev_event ( struct notifier_block * this , unsigned long event , void * ptr )
{
2013-05-28 05:30:21 +04:00
struct net_device * dev = netdev_notifier_info_to_dev ( ptr ) ;
2015-12-10 21:25:24 +03:00
struct netdev_notifier_changeupper_info * info ;
2012-08-22 21:19:46 +04:00
struct in_device * in_dev ;
2011-03-25 03:42:21 +03:00
struct net * net = dev_net ( dev ) ;
2015-06-23 20:45:36 +03:00
unsigned int flags ;
2005-04-17 02:20:36 +04:00
if ( event = = NETDEV_UNREGISTER ) {
2015-10-30 11:23:33 +03:00
fib_disable_ip ( dev , event , true ) ;
2012-08-01 02:06:50 +04:00
rt_flush_dev ( dev ) ;
2005-04-17 02:20:36 +04:00
return NOTIFY_DONE ;
}
2012-08-22 21:19:46 +04:00
in_dev = __in_dev_get_rtnl ( dev ) ;
2014-01-23 13:19:34 +04:00
if ( ! in_dev )
return NOTIFY_DONE ;
2012-08-22 21:19:46 +04:00
2005-04-17 02:20:36 +04:00
switch ( event ) {
case NETDEV_UP :
for_ifa ( in_dev ) {
fib_add_ifaddr ( ifa ) ;
} endfor_ifa ( in_dev ) ;
# ifdef CONFIG_IP_ROUTE_MULTIPATH
2015-06-23 20:45:36 +03:00
fib_sync_up ( dev , RTNH_F_DEAD ) ;
2005-04-17 02:20:36 +04:00
# endif
2011-03-25 03:42:21 +03:00
atomic_inc ( & net - > ipv4 . dev_addr_genid ) ;
2012-09-07 04:45:29 +04:00
rt_cache_flush ( net ) ;
2005-04-17 02:20:36 +04:00
break ;
case NETDEV_DOWN :
2015-10-30 11:23:33 +03:00
fib_disable_ip ( dev , event , false ) ;
2005-04-17 02:20:36 +04:00
break ;
case NETDEV_CHANGE :
2015-06-23 20:45:36 +03:00
flags = dev_get_flags ( dev ) ;
if ( flags & ( IFF_RUNNING | IFF_LOWER_UP ) )
fib_sync_up ( dev , RTNH_F_LINKDOWN ) ;
else
2015-10-30 11:23:33 +03:00
fib_sync_down_dev ( dev , event , false ) ;
2015-06-23 20:45:36 +03:00
/* fall through */
case NETDEV_CHANGEMTU :
2012-09-07 04:45:29 +04:00
rt_cache_flush ( net ) ;
2005-04-17 02:20:36 +04:00
break ;
2015-12-10 21:25:24 +03:00
case NETDEV_CHANGEUPPER :
info = ptr ;
/* flush all routes if dev is linked to or unlinked from
* an L3 master device ( e . g . , VRF )
*/
if ( info - > upper_dev & & netif_is_l3_master ( info - > upper_dev ) )
fib_disable_ip ( dev , NETDEV_DOWN , true ) ;
break ;
2005-04-17 02:20:36 +04:00
}
return NOTIFY_DONE ;
}
static struct notifier_block fib_inetaddr_notifier = {
2008-11-03 11:25:16 +03:00
. notifier_call = fib_inetaddr_event ,
2005-04-17 02:20:36 +04:00
} ;
static struct notifier_block fib_netdev_notifier = {
2008-11-03 11:25:16 +03:00
. notifier_call = fib_netdev_event ,
2005-04-17 02:20:36 +04:00
} ;
2008-01-10 14:22:17 +03:00
static int __net_init ip_fib_net_init ( struct net * net )
2005-04-17 02:20:36 +04:00
{
2008-02-01 05:44:53 +03:00
int err ;
2010-10-13 12:22:03 +04:00
size_t size = sizeof ( struct hlist_head ) * FIB_TABLE_HASHSZ ;
2016-12-03 18:45:06 +03:00
net - > ipv4 . fib_seq = 0 ;
2010-10-13 12:22:03 +04:00
/* Avoid false sharing : Use at least a full cache line */
size = max_t ( size_t , size , L1_CACHE_BYTES ) ;
2006-08-11 10:10:46 +04:00
2010-10-13 12:22:03 +04:00
net - > ipv4 . fib_table_hash = kzalloc ( size , GFP_KERNEL ) ;
2015-04-03 11:17:26 +03:00
if ( ! net - > ipv4 . fib_table_hash )
2008-01-10 14:28:24 +03:00
return - ENOMEM ;
2008-02-01 05:44:53 +03:00
err = fib4_rules_init ( net ) ;
if ( err < 0 )
goto fail ;
return 0 ;
fail :
kfree ( net - > ipv4 . fib_table_hash ) ;
return err ;
2008-01-10 14:22:17 +03:00
}
2005-04-17 02:20:36 +04:00
2010-01-17 06:35:32 +03:00
static void ip_fib_net_exit ( struct net * net )
2008-01-10 14:22:17 +03:00
{
unsigned int i ;
2015-03-10 23:03:54 +03:00
rtnl_lock ( ) ;
2015-03-28 00:14:22 +03:00
# ifdef CONFIG_IP_MULTIPLE_TABLES
RCU_INIT_POINTER ( net - > ipv4 . fib_main , NULL ) ;
RCU_INIT_POINTER ( net - > ipv4 . fib_default , NULL ) ;
# endif
2008-01-10 14:22:17 +03:00
for ( i = 0 ; i < FIB_TABLE_HASHSZ ; i + + ) {
2015-03-05 02:02:44 +03:00
struct hlist_head * head = & net - > ipv4 . fib_table_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>
2013-02-28 05:06:00 +04:00
struct hlist_node * tmp ;
2015-03-05 02:02:44 +03:00
struct fib_table * tb ;
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>
2013-02-28 05:06:00 +04:00
hlist_for_each_entry_safe ( tb , tmp , head , tb_hlist ) {
hlist_del ( & tb - > tb_hlist ) ;
2016-09-26 13:52:29 +03:00
fib_table_flush ( net , tb ) ;
2010-10-28 06:00:43 +04:00
fib_free_table ( tb ) ;
2008-01-10 14:22:17 +03:00
}
}
2015-03-28 00:14:16 +03:00
# ifdef CONFIG_IP_MULTIPLE_TABLES
fib4_rules_exit ( net ) ;
# endif
2011-03-31 03:57:46 +04:00
rtnl_unlock ( ) ;
2008-01-10 14:28:24 +03:00
kfree ( net - > ipv4 . fib_table_hash ) ;
2008-01-10 14:22:17 +03:00
}
static int __net_init fib_net_init ( struct net * net )
{
int error ;
2012-07-06 09:13:13 +04:00
# ifdef CONFIG_IP_ROUTE_CLASSID
net - > ipv4 . fib_num_tclassid_users = 0 ;
# endif
2008-01-10 14:22:17 +03:00
error = ip_fib_net_init ( net ) ;
if ( error < 0 )
goto out ;
error = nl_fib_lookup_init ( net ) ;
if ( error < 0 )
goto out_nlfl ;
error = fib_proc_init ( net ) ;
if ( error < 0 )
goto out_proc ;
out :
return error ;
out_proc :
nl_fib_lookup_exit ( net ) ;
out_nlfl :
ip_fib_net_exit ( net ) ;
goto out ;
}
static void __net_exit fib_net_exit ( struct net * net )
{
fib_proc_exit ( net ) ;
nl_fib_lookup_exit ( net ) ;
ip_fib_net_exit ( net ) ;
}
static struct pernet_operations fib_net_ops = {
. init = fib_net_init ,
. exit = fib_net_exit ,
} ;
void __init ip_fib_init ( void )
{
2011-06-10 05:27:09 +04:00
rtnl_register ( PF_INET , RTM_NEWROUTE , inet_rtm_newroute , NULL , NULL ) ;
rtnl_register ( PF_INET , RTM_DELROUTE , inet_rtm_delroute , NULL , NULL ) ;
rtnl_register ( PF_INET , RTM_GETROUTE , NULL , inet_dump_fib , NULL ) ;
2008-01-10 14:22:17 +03:00
register_pernet_subsys ( & fib_net_ops ) ;
register_netdevice_notifier ( & fib_netdev_notifier ) ;
register_inetaddr_notifier ( & fib_inetaddr_notifier ) ;
2008-01-15 10:14:20 +03:00
2011-02-02 02:30:56 +03:00
fib_trie_init ( ) ;
2005-04-17 02:20:36 +04:00
}