2019-05-27 06:55:01 +00:00
|
|
|
/* SPDX-License-Identifier: GPL-2.0-or-later */
|
2005-08-10 02:59:20 +00:00
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*
|
|
|
|
* Authors: Lotsa people, from code originally in tcp
|
|
|
|
*/
|
|
|
|
|
|
|
|
#ifndef _INET_HASHTABLES_H
|
|
|
|
#define _INET_HASHTABLES_H
|
|
|
|
|
2005-08-10 03:09:30 +00:00
|
|
|
|
2005-08-10 03:07:13 +00:00
|
|
|
#include <linux/interrupt.h>
|
2008-10-07 18:41:57 +00:00
|
|
|
#include <linux/ip.h>
|
2005-08-10 03:09:06 +00:00
|
|
|
#include <linux/ipv6.h>
|
2005-08-10 03:00:51 +00:00
|
|
|
#include <linux/list.h>
|
|
|
|
#include <linux/slab.h>
|
2005-08-10 03:09:06 +00:00
|
|
|
#include <linux/socket.h>
|
2005-08-10 03:00:51 +00:00
|
|
|
#include <linux/spinlock.h>
|
2005-08-10 02:59:20 +00:00
|
|
|
#include <linux/types.h>
|
2005-08-10 03:08:09 +00:00
|
|
|
#include <linux/wait.h>
|
2005-08-10 02:59:20 +00:00
|
|
|
|
2005-08-10 03:10:42 +00:00
|
|
|
#include <net/inet_connection_sock.h>
|
2005-12-27 04:43:12 +00:00
|
|
|
#include <net/inet_sock.h>
|
2005-08-10 03:07:13 +00:00
|
|
|
#include <net/sock.h>
|
2008-10-07 18:41:57 +00:00
|
|
|
#include <net/route.h>
|
2005-08-10 03:08:28 +00:00
|
|
|
#include <net/tcp_states.h>
|
2008-06-17 00:14:11 +00:00
|
|
|
#include <net/netns/hash.h>
|
2005-08-10 03:07:13 +00:00
|
|
|
|
2017-06-30 10:08:01 +00:00
|
|
|
#include <linux/refcount.h>
|
2005-08-10 03:09:46 +00:00
|
|
|
#include <asm/byteorder.h>
|
2005-08-10 03:08:09 +00:00
|
|
|
|
2005-08-10 03:00:51 +00:00
|
|
|
/* This is for all connections with a full identity, no wildcards.
|
tcp/dccp: remove twchain
TCP listener refactoring, part 3 :
Our goal is to hash SYN_RECV sockets into main ehash for fast lookup,
and parallel SYN processing.
Current inet_ehash_bucket contains two chains, one for ESTABLISH (and
friend states) sockets, another for TIME_WAIT sockets only.
As the hash table is sized to get at most one socket per bucket, it
makes little sense to have separate twchain, as it makes the lookup
slightly more complicated, and doubles hash table memory usage.
If we make sure all socket types have the lookup keys at the same
offsets, we can use a generic and faster lookup. It turns out TIME_WAIT
and ESTABLISHED sockets already have common lookup fields for IPv4.
[ INET_TW_MATCH() is no longer needed ]
I'll provide a follow-up to factorize IPv6 lookup as well, to remove
INET6_TW_MATCH()
This way, SYN_RECV pseudo sockets will be supported the same.
A new sock_gen_put() helper is added, doing either a sock_put() or
inet_twsk_put() [ and will support SYN_RECV later ].
Note this helper should only be called in real slow path, when rcu
lookup found a socket that was moved to another identity (freed/reused
immediately), but could eventually be used in other contexts, like
sock_edemux()
Before patch :
dmesg | grep "TCP established"
TCP established hash table entries: 524288 (order: 11, 8388608 bytes)
After patch :
TCP established hash table entries: 524288 (order: 10, 4194304 bytes)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-03 07:22:02 +00:00
|
|
|
* The 'e' prefix stands for Establish, but we really put all sockets
|
|
|
|
* but LISTEN ones.
|
2005-08-10 03:00:51 +00:00
|
|
|
*/
|
|
|
|
struct inet_ehash_bucket {
|
2008-11-17 03:40:17 +00:00
|
|
|
struct hlist_nulls_head chain;
|
2005-10-04 22:55:51 +00:00
|
|
|
};
|
2005-08-10 03:00:51 +00:00
|
|
|
|
|
|
|
/* There are a few simple rules, which allow for local port reuse by
|
|
|
|
* an application. In essence:
|
|
|
|
*
|
|
|
|
* 1) Sockets bound to different interfaces may share a local port.
|
|
|
|
* Failing that, goto test 2.
|
|
|
|
* 2) If all sockets have sk->sk_reuse set, and none of them are in
|
|
|
|
* TCP_LISTEN state, the port may be shared.
|
|
|
|
* Failing that, goto test 3.
|
|
|
|
* 3) If all sockets are bound to a specific inet_sk(sk)->rcv_saddr local
|
|
|
|
* address, and none of them are the same, the port may be
|
|
|
|
* shared.
|
|
|
|
* Failing this, the port cannot be shared.
|
|
|
|
*
|
|
|
|
* The interesting point, is test #2. This is what an FTP server does
|
|
|
|
* all day. To optimize this case we use a specific flag bit defined
|
|
|
|
* below. As we add sockets to a bind bucket list, we perform a
|
|
|
|
* check of: (newsk->sk_reuse && (newsk->sk_state != TCP_LISTEN))
|
|
|
|
* As long as all sockets added to a bind bucket pass this test,
|
|
|
|
* the flag bit will be set.
|
|
|
|
* The resulting situation is that tcp_v[46]_verify_bind() can just check
|
|
|
|
* for this flag bit, if it is set and the socket trying to bind has
|
|
|
|
* sk->sk_reuse set, we don't even have to walk the owners list at all,
|
|
|
|
* we return that it is ok to bind this socket to the requested local port.
|
|
|
|
*
|
|
|
|
* Sounds like a lot of work, but it is worth it. In a more naive
|
|
|
|
* implementation (ie. current FreeBSD etc.) the entire list of ports
|
|
|
|
* must be walked for each data port opened by an ftp server. Needless
|
|
|
|
* to say, this does not scale at all. With a couple thousand FTP
|
|
|
|
* users logged onto your box, isn't it nice to know that new data
|
|
|
|
* ports are created in O(1) time? I thought so. ;-) -DaveM
|
|
|
|
*/
|
2017-01-17 15:51:06 +00:00
|
|
|
#define FASTREUSEPORT_ANY 1
|
|
|
|
#define FASTREUSEPORT_STRICT 2
|
|
|
|
|
2005-08-10 03:00:51 +00:00
|
|
|
struct inet_bind_bucket {
|
2015-03-12 04:06:44 +00:00
|
|
|
possible_net_t ib_net;
|
2018-11-07 15:36:02 +00:00
|
|
|
int l3mdev;
|
2005-08-10 03:00:51 +00:00
|
|
|
unsigned short port;
|
2013-01-22 09:50:24 +00:00
|
|
|
signed char fastreuse;
|
|
|
|
signed char fastreuseport;
|
|
|
|
kuid_t fastuid;
|
2017-01-17 15:51:06 +00:00
|
|
|
#if IS_ENABLED(CONFIG_IPV6)
|
|
|
|
struct in6_addr fast_v6_rcv_saddr;
|
|
|
|
#endif
|
|
|
|
__be32 fast_rcv_saddr;
|
|
|
|
unsigned short fast_sk_family;
|
|
|
|
bool fast_ipv6_only;
|
2005-08-10 03:00:51 +00:00
|
|
|
struct hlist_node node;
|
|
|
|
struct hlist_head owners;
|
|
|
|
};
|
|
|
|
|
2008-11-12 08:54:20 +00:00
|
|
|
static inline struct net *ib_net(struct inet_bind_bucket *ib)
|
|
|
|
{
|
|
|
|
return read_pnet(&ib->ib_net);
|
|
|
|
}
|
|
|
|
|
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 01:06:00 +00:00
|
|
|
#define inet_bind_bucket_for_each(tb, head) \
|
|
|
|
hlist_for_each_entry(tb, head, node)
|
2005-08-10 03:00:51 +00:00
|
|
|
|
|
|
|
struct inet_bind_hashbucket {
|
|
|
|
spinlock_t lock;
|
|
|
|
struct hlist_head chain;
|
|
|
|
};
|
|
|
|
|
2019-12-14 02:20:41 +00:00
|
|
|
/* Sockets can be hashed in established or listening table.
|
|
|
|
* We must use different 'nulls' end-of-chain value for all hash buckets :
|
|
|
|
* A socket might transition from ESTABLISH to LISTEN state without
|
|
|
|
* RCU grace period. A lookup in ehash table needs to handle this case.
|
2008-11-24 01:22:55 +00:00
|
|
|
*/
|
2019-12-14 02:20:41 +00:00
|
|
|
#define LISTENING_NULLS_BASE (1U << 29)
|
2008-11-20 08:40:07 +00:00
|
|
|
struct inet_listen_hashbucket {
|
|
|
|
spinlock_t lock;
|
2022-05-12 00:06:05 +00:00
|
|
|
struct hlist_nulls_head nulls_head;
|
2008-11-20 08:40:07 +00:00
|
|
|
};
|
|
|
|
|
2005-08-10 03:00:51 +00:00
|
|
|
/* This is for listening sockets, thus all sockets which possess wildcards. */
|
|
|
|
#define INET_LHTABLE_SIZE 32 /* Yes, really, this is all you need. */
|
|
|
|
|
|
|
|
struct inet_hashinfo {
|
|
|
|
/* This is for sockets with full identity only. Sockets here will
|
|
|
|
* always be without wildcards and will have the following invariant:
|
|
|
|
*
|
|
|
|
* TCP_ESTABLISHED <= sk->sk_state < TCP_CLOSE
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
struct inet_ehash_bucket *ehash;
|
2008-11-21 04:39:09 +00:00
|
|
|
spinlock_t *ehash_locks;
|
2009-10-09 00:16:19 +00:00
|
|
|
unsigned int ehash_mask;
|
2007-11-07 10:40:20 +00:00
|
|
|
unsigned int ehash_locks_mask;
|
2005-08-10 03:00:51 +00:00
|
|
|
|
|
|
|
/* Ok, let's try this, I give up, we do need a local binding
|
|
|
|
* TCP hash as well as the others for fast bind/connect.
|
|
|
|
*/
|
2017-12-01 20:52:31 +00:00
|
|
|
struct kmem_cache *bind_bucket_cachep;
|
2005-08-10 03:00:51 +00:00
|
|
|
struct inet_bind_hashbucket *bhash;
|
2007-10-09 08:59:42 +00:00
|
|
|
unsigned int bhash_size;
|
2005-08-10 03:00:51 +00:00
|
|
|
|
2017-12-01 20:52:31 +00:00
|
|
|
/* The 2nd listener table hashed by local port and address */
|
|
|
|
unsigned int lhash2_mask;
|
|
|
|
struct inet_listen_hashbucket *lhash2;
|
2005-08-10 03:00:51 +00:00
|
|
|
};
|
|
|
|
|
2017-12-01 20:52:31 +00:00
|
|
|
static inline struct inet_listen_hashbucket *
|
|
|
|
inet_lhash2_bucket(struct inet_hashinfo *h, u32 hash)
|
|
|
|
{
|
|
|
|
return &h->lhash2[hash & h->lhash2_mask];
|
|
|
|
}
|
|
|
|
|
[INET]: speedup inet (tcp/dccp) lookups
Arnaldo and I agreed it could be applied now, because I have other
pending patches depending on this one (Thank you Arnaldo)
(The other important patch moves skc_refcnt in a separate cache line,
so that the SMP/NUMA performance doesnt suffer from cache line ping pongs)
1) First some performance data :
--------------------------------
tcp_v4_rcv() wastes a *lot* of time in __inet_lookup_established()
The most time critical code is :
sk_for_each(sk, node, &head->chain) {
if (INET_MATCH(sk, acookie, saddr, daddr, ports, dif))
goto hit; /* You sunk my battleship! */
}
The sk_for_each() does use prefetch() hints but only the begining of
"struct sock" is prefetched.
As INET_MATCH first comparison uses inet_sk(__sk)->daddr, wich is far
away from the begining of "struct sock", it has to bring into CPU
cache cold cache line. Each iteration has to use at least 2 cache
lines.
This can be problematic if some chains are very long.
2) The goal
-----------
The idea I had is to change things so that INET_MATCH() may return
FALSE in 99% of cases only using the data already in the CPU cache,
using one cache line per iteration.
3) Description of the patch
---------------------------
Adds a new 'unsigned int skc_hash' field in 'struct sock_common',
filling a 32 bits hole on 64 bits platform.
struct sock_common {
unsigned short skc_family;
volatile unsigned char skc_state;
unsigned char skc_reuse;
int skc_bound_dev_if;
struct hlist_node skc_node;
struct hlist_node skc_bind_node;
atomic_t skc_refcnt;
+ unsigned int skc_hash;
struct proto *skc_prot;
};
Store in this 32 bits field the full hash, not masked by (ehash_size -
1) Using this full hash as the first comparison done in INET_MATCH
permits us immediatly skip the element without touching a second cache
line in case of a miss.
Suppress the sk_hashent/tw_hashent fields since skc_hash (aliased to
sk_hash and tw_hash) already contains the slot number if we mask with
(ehash_size - 1)
File include/net/inet_hashtables.h
64 bits platforms :
#define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
(((__sk)->sk_hash == (__hash))
((*((__u64 *)&(inet_sk(__sk)->daddr)))== (__cookie)) && \
((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \
(!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
32bits platforms:
#define TCP_IPV4_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
(((__sk)->sk_hash == (__hash)) && \
(inet_sk(__sk)->daddr == (__saddr)) && \
(inet_sk(__sk)->rcv_saddr == (__daddr)) && \
(!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
- Adds a prefetch(head->chain.first) in
__inet_lookup_established()/__tcp_v4_check_established() and
__inet6_lookup_established()/__tcp_v6_check_established() and
__dccp_v4_check_established() to bring into cache the first element of the
list, before the {read|write}_lock(&head->lock);
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2005-10-03 21:13:38 +00:00
|
|
|
static inline struct inet_ehash_bucket *inet_ehash_bucket(
|
|
|
|
struct inet_hashinfo *hashinfo,
|
|
|
|
unsigned int hash)
|
|
|
|
{
|
2009-10-09 00:16:19 +00:00
|
|
|
return &hashinfo->ehash[hash & hashinfo->ehash_mask];
|
2005-08-10 02:59:20 +00:00
|
|
|
}
|
|
|
|
|
2008-11-21 04:39:09 +00:00
|
|
|
static inline spinlock_t *inet_ehash_lockp(
|
2007-11-07 10:40:20 +00:00
|
|
|
struct inet_hashinfo *hashinfo,
|
|
|
|
unsigned int hash)
|
|
|
|
{
|
|
|
|
return &hashinfo->ehash_locks[hash & hashinfo->ehash_locks_mask];
|
|
|
|
}
|
|
|
|
|
2015-05-26 14:55:34 +00:00
|
|
|
int inet_ehash_locks_alloc(struct inet_hashinfo *hashinfo);
|
2007-11-07 10:40:20 +00:00
|
|
|
|
2020-06-09 14:18:16 +00:00
|
|
|
static inline void inet_hashinfo2_free_mod(struct inet_hashinfo *h)
|
|
|
|
{
|
|
|
|
kfree(h->lhash2);
|
|
|
|
h->lhash2 = NULL;
|
|
|
|
}
|
|
|
|
|
2007-11-07 10:40:20 +00:00
|
|
|
static inline void inet_ehash_locks_free(struct inet_hashinfo *hashinfo)
|
|
|
|
{
|
2015-05-26 14:55:34 +00:00
|
|
|
kvfree(hashinfo->ehash_locks);
|
|
|
|
hashinfo->ehash_locks = NULL;
|
2007-11-07 10:40:20 +00:00
|
|
|
}
|
|
|
|
|
2018-11-07 15:36:03 +00:00
|
|
|
static inline bool inet_sk_bound_dev_eq(struct net *net, int bound_dev_if,
|
|
|
|
int dif, int sdif)
|
|
|
|
{
|
|
|
|
#if IS_ENABLED(CONFIG_NET_L3_MASTER_DEV)
|
2022-07-13 20:51:59 +00:00
|
|
|
return inet_bound_dev_eq(!!READ_ONCE(net->ipv4.sysctl_tcp_l3mdev_accept),
|
2018-11-07 15:36:03 +00:00
|
|
|
bound_dev_if, dif, sdif);
|
|
|
|
#else
|
|
|
|
return inet_bound_dev_eq(true, bound_dev_if, dif, sdif);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2013-09-21 17:22:41 +00:00
|
|
|
struct inet_bind_bucket *
|
|
|
|
inet_bind_bucket_create(struct kmem_cache *cachep, struct net *net,
|
|
|
|
struct inet_bind_hashbucket *head,
|
2018-11-07 15:36:02 +00:00
|
|
|
const unsigned short snum, int l3mdev);
|
2013-09-21 17:22:41 +00:00
|
|
|
void inet_bind_bucket_destroy(struct kmem_cache *cachep,
|
|
|
|
struct inet_bind_bucket *tb);
|
|
|
|
|
2015-03-18 21:05:33 +00:00
|
|
|
static inline u32 inet_bhashfn(const struct net *net, const __u16 lport,
|
|
|
|
const u32 bhash_size)
|
2005-08-10 03:00:51 +00:00
|
|
|
{
|
2008-06-17 00:14:11 +00:00
|
|
|
return (lport + net_hash_mix(net)) & (bhash_size - 1);
|
2005-08-10 03:00:51 +00:00
|
|
|
}
|
|
|
|
|
2013-09-21 17:22:41 +00:00
|
|
|
void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
|
2022-06-15 19:32:13 +00:00
|
|
|
const unsigned short snum);
|
2005-08-10 03:07:13 +00:00
|
|
|
|
|
|
|
/* Caller must disable local BH processing. */
|
2015-09-29 14:42:44 +00:00
|
|
|
int __inet_inherit_port(const struct sock *sk, struct sock *child);
|
2005-08-10 03:07:13 +00:00
|
|
|
|
2013-09-21 17:22:41 +00:00
|
|
|
void inet_put_port(struct sock *sk);
|
2005-08-10 03:07:13 +00:00
|
|
|
|
2017-12-01 20:52:31 +00:00
|
|
|
void inet_hashinfo2_init(struct inet_hashinfo *h, const char *name,
|
|
|
|
unsigned long numentries, int scale,
|
|
|
|
unsigned long low_limit,
|
|
|
|
unsigned long high_limit);
|
2018-12-24 20:57:17 +00:00
|
|
|
int inet_hashinfo2_init_mod(struct inet_hashinfo *h);
|
2005-08-10 03:08:09 +00:00
|
|
|
|
tcp: fix race condition when creating child sockets from syncookies
When the TCP stack is in SYN flood mode, the server child socket is
created from the SYN cookie received in a TCP packet with the ACK flag
set.
The child socket is created when the server receives the first TCP
packet with a valid SYN cookie from the client. Usually, this packet
corresponds to the final step of the TCP 3-way handshake, the ACK
packet. But is also possible to receive a valid SYN cookie from the
first TCP data packet sent by the client, and thus create a child socket
from that SYN cookie.
Since a client socket is ready to send data as soon as it receives the
SYN+ACK packet from the server, the client can send the ACK packet (sent
by the TCP stack code), and the first data packet (sent by the userspace
program) almost at the same time, and thus the server will equally
receive the two TCP packets with valid SYN cookies almost at the same
instant.
When such event happens, the TCP stack code has a race condition that
occurs between the momement a lookup is done to the established
connections hashtable to check for the existence of a connection for the
same client, and the moment that the child socket is added to the
established connections hashtable. As a consequence, this race condition
can lead to a situation where we add two child sockets to the
established connections hashtable and deliver two sockets to the
userspace program to the same client.
This patch fixes the race condition by checking if an existing child
socket exists for the same client when we are adding the second child
socket to the established connections socket. If an existing child
socket exists, we drop the packet and discard the second child socket
to the same client.
Signed-off-by: Ricardo Dias <rdias@singlestore.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20201120111133.GA67501@rdias-suse-pc.lan
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-11-20 11:11:33 +00:00
|
|
|
bool inet_ehash_insert(struct sock *sk, struct sock *osk, bool *found_dup_sk);
|
|
|
|
bool inet_ehash_nolisten(struct sock *sk, struct sock *osk,
|
|
|
|
bool *found_dup_sk);
|
2017-01-17 15:51:01 +00:00
|
|
|
int __inet_hash(struct sock *sk, struct sock *osk);
|
2016-02-10 16:50:35 +00:00
|
|
|
int inet_hash(struct sock *sk);
|
2013-09-21 17:22:41 +00:00
|
|
|
void inet_unhash(struct sock *sk);
|
2005-08-10 03:09:06 +00:00
|
|
|
|
2013-09-21 17:22:41 +00:00
|
|
|
struct sock *__inet_lookup_listener(struct net *net,
|
|
|
|
struct inet_hashinfo *hashinfo,
|
2016-02-10 16:50:38 +00:00
|
|
|
struct sk_buff *skb, int doff,
|
2013-09-21 17:22:41 +00:00
|
|
|
const __be32 saddr, const __be16 sport,
|
|
|
|
const __be32 daddr,
|
|
|
|
const unsigned short hnum,
|
2017-08-07 15:44:17 +00:00
|
|
|
const int dif, const int sdif);
|
2006-08-09 22:47:12 +00:00
|
|
|
|
2008-01-31 13:06:40 +00:00
|
|
|
static inline struct sock *inet_lookup_listener(struct net *net,
|
|
|
|
struct inet_hashinfo *hashinfo,
|
2016-02-10 16:50:38 +00:00
|
|
|
struct sk_buff *skb, int doff,
|
2013-01-22 09:50:24 +00:00
|
|
|
__be32 saddr, __be16 sport,
|
2017-08-07 15:44:17 +00:00
|
|
|
__be32 daddr, __be16 dport, int dif, int sdif)
|
2006-08-09 22:47:12 +00:00
|
|
|
{
|
2016-02-10 16:50:38 +00:00
|
|
|
return __inet_lookup_listener(net, hashinfo, skb, doff, saddr, sport,
|
2017-08-07 15:44:17 +00:00
|
|
|
daddr, ntohs(dport), dif, sdif);
|
2006-08-09 22:47:12 +00:00
|
|
|
}
|
2005-08-10 03:09:30 +00:00
|
|
|
|
|
|
|
/* Socket demux engine toys. */
|
2006-09-28 01:43:07 +00:00
|
|
|
/* What happens here is ugly; there's a pair of adjacent fields in
|
|
|
|
struct inet_sock; __be16 dport followed by __u16 num. We want to
|
|
|
|
search by pair, so we combine the keys into a single 32bit value
|
|
|
|
and compare with 32bit value read from &...->dport. Let's at least
|
|
|
|
make sure that it's not mixed with anything else...
|
|
|
|
On 64bit targets we combine comparisons with pair of adjacent __be32
|
|
|
|
fields in the same way.
|
|
|
|
*/
|
2005-08-10 03:09:30 +00:00
|
|
|
#ifdef __BIG_ENDIAN
|
|
|
|
#define INET_COMBINED_PORTS(__sport, __dport) \
|
2006-09-28 01:43:07 +00:00
|
|
|
((__force __portpair)(((__force __u32)(__be16)(__sport) << 16) | (__u32)(__dport)))
|
2005-08-10 03:09:30 +00:00
|
|
|
#else /* __LITTLE_ENDIAN */
|
|
|
|
#define INET_COMBINED_PORTS(__sport, __dport) \
|
2006-09-28 01:43:07 +00:00
|
|
|
((__force __portpair)(((__u32)(__dport) << 16) | (__force __u32)(__be16)(__sport)))
|
2005-08-10 03:09:30 +00:00
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifdef __BIG_ENDIAN
|
|
|
|
#define INET_ADDR_COOKIE(__name, __saddr, __daddr) \
|
2006-09-28 01:43:07 +00:00
|
|
|
const __addrpair __name = (__force __addrpair) ( \
|
|
|
|
(((__force __u64)(__be32)(__saddr)) << 32) | \
|
2014-05-14 03:30:07 +00:00
|
|
|
((__force __u64)(__be32)(__daddr)))
|
2005-08-10 03:09:30 +00:00
|
|
|
#else /* __LITTLE_ENDIAN */
|
|
|
|
#define INET_ADDR_COOKIE(__name, __saddr, __daddr) \
|
2006-09-28 01:43:07 +00:00
|
|
|
const __addrpair __name = (__force __addrpair) ( \
|
|
|
|
(((__force __u64)(__be32)(__daddr)) << 32) | \
|
2014-05-14 03:30:07 +00:00
|
|
|
((__force __u64)(__be32)(__saddr)))
|
2005-08-10 03:09:30 +00:00
|
|
|
#endif /* __BIG_ENDIAN */
|
2022-05-12 16:56:01 +00:00
|
|
|
|
2022-05-13 18:55:50 +00:00
|
|
|
static inline bool inet_match(struct net *net, const struct sock *sk,
|
2022-05-12 16:56:01 +00:00
|
|
|
const __addrpair cookie, const __portpair ports,
|
|
|
|
int dif, int sdif)
|
|
|
|
{
|
|
|
|
int bound_dev_if;
|
|
|
|
|
|
|
|
if (!net_eq(sock_net(sk), net) ||
|
|
|
|
sk->sk_portpair != ports ||
|
|
|
|
sk->sk_addrpair != cookie)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
/* Paired with WRITE_ONCE() from sock_bindtoindex_locked() */
|
|
|
|
bound_dev_if = READ_ONCE(sk->sk_bound_dev_if);
|
|
|
|
return bound_dev_if == dif || bound_dev_if == sdif;
|
|
|
|
}
|
2005-08-10 03:09:46 +00:00
|
|
|
|
2016-04-01 15:52:14 +00:00
|
|
|
/* Sockets in TCP_CLOSE state are _always_ taken out of the hash, so we need
|
2005-08-10 03:09:46 +00:00
|
|
|
* not check it for lookups anymore, thanks Alexey. -DaveM
|
|
|
|
*/
|
2013-09-21 17:22:41 +00:00
|
|
|
struct sock *__inet_lookup_established(struct net *net,
|
|
|
|
struct inet_hashinfo *hashinfo,
|
|
|
|
const __be32 saddr, const __be16 sport,
|
|
|
|
const __be32 daddr, const u16 hnum,
|
2017-08-07 15:44:17 +00:00
|
|
|
const int dif, const int sdif);
|
2005-08-10 03:09:46 +00:00
|
|
|
|
2006-08-09 22:47:12 +00:00
|
|
|
static inline struct sock *
|
2008-01-31 13:06:40 +00:00
|
|
|
inet_lookup_established(struct net *net, struct inet_hashinfo *hashinfo,
|
2006-09-28 01:43:33 +00:00
|
|
|
const __be32 saddr, const __be16 sport,
|
|
|
|
const __be32 daddr, const __be16 dport,
|
2006-08-09 22:47:12 +00:00
|
|
|
const int dif)
|
|
|
|
{
|
2008-01-31 13:06:40 +00:00
|
|
|
return __inet_lookup_established(net, hashinfo, saddr, sport, daddr,
|
2017-08-07 15:44:17 +00:00
|
|
|
ntohs(dport), dif, 0);
|
2006-08-09 22:47:12 +00:00
|
|
|
}
|
|
|
|
|
2008-01-31 13:06:40 +00:00
|
|
|
static inline struct sock *__inet_lookup(struct net *net,
|
|
|
|
struct inet_hashinfo *hashinfo,
|
2016-02-10 16:50:38 +00:00
|
|
|
struct sk_buff *skb, int doff,
|
2006-09-28 01:43:33 +00:00
|
|
|
const __be32 saddr, const __be16 sport,
|
|
|
|
const __be32 daddr, const __be16 dport,
|
2017-08-07 15:44:17 +00:00
|
|
|
const int dif, const int sdif,
|
2016-04-01 15:52:17 +00:00
|
|
|
bool *refcounted)
|
2005-08-10 03:09:46 +00:00
|
|
|
{
|
2006-08-09 22:47:12 +00:00
|
|
|
u16 hnum = ntohs(dport);
|
2016-04-01 15:52:17 +00:00
|
|
|
struct sock *sk;
|
2008-01-31 13:06:40 +00:00
|
|
|
|
2016-04-01 15:52:17 +00:00
|
|
|
sk = __inet_lookup_established(net, hashinfo, saddr, sport,
|
2017-08-07 15:44:17 +00:00
|
|
|
daddr, hnum, dif, sdif);
|
2016-04-01 15:52:17 +00:00
|
|
|
*refcounted = true;
|
|
|
|
if (sk)
|
|
|
|
return sk;
|
|
|
|
*refcounted = false;
|
|
|
|
return __inet_lookup_listener(net, hashinfo, skb, doff, saddr,
|
2017-08-07 15:44:17 +00:00
|
|
|
sport, daddr, hnum, dif, sdif);
|
2005-08-10 03:09:46 +00:00
|
|
|
}
|
|
|
|
|
2008-01-31 13:06:40 +00:00
|
|
|
static inline struct sock *inet_lookup(struct net *net,
|
|
|
|
struct inet_hashinfo *hashinfo,
|
2016-02-10 16:50:38 +00:00
|
|
|
struct sk_buff *skb, int doff,
|
2006-09-28 01:43:33 +00:00
|
|
|
const __be32 saddr, const __be16 sport,
|
|
|
|
const __be32 daddr, const __be16 dport,
|
2005-08-10 03:09:46 +00:00
|
|
|
const int dif)
|
|
|
|
{
|
|
|
|
struct sock *sk;
|
2016-04-01 15:52:17 +00:00
|
|
|
bool refcounted;
|
2005-08-10 03:09:46 +00:00
|
|
|
|
2016-02-10 16:50:38 +00:00
|
|
|
sk = __inet_lookup(net, hashinfo, skb, doff, saddr, sport, daddr,
|
2017-08-07 15:44:17 +00:00
|
|
|
dport, dif, 0, &refcounted);
|
2005-08-10 03:09:46 +00:00
|
|
|
|
2017-06-30 10:08:01 +00:00
|
|
|
if (sk && !refcounted && !refcount_inc_not_zero(&sk->sk_refcnt))
|
2016-04-01 15:52:17 +00:00
|
|
|
sk = NULL;
|
2005-08-10 03:09:46 +00:00
|
|
|
return sk;
|
|
|
|
}
|
2005-12-14 07:25:31 +00:00
|
|
|
|
2008-10-07 18:41:57 +00:00
|
|
|
static inline struct sock *__inet_lookup_skb(struct inet_hashinfo *hashinfo,
|
|
|
|
struct sk_buff *skb,
|
2016-02-10 16:50:38 +00:00
|
|
|
int doff,
|
2008-10-07 18:41:57 +00:00
|
|
|
const __be16 sport,
|
2016-04-01 15:52:17 +00:00
|
|
|
const __be16 dport,
|
2017-08-07 15:44:17 +00:00
|
|
|
const int sdif,
|
2016-04-01 15:52:17 +00:00
|
|
|
bool *refcounted)
|
2008-10-07 18:41:57 +00:00
|
|
|
{
|
2020-03-29 22:53:39 +00:00
|
|
|
struct sock *sk = skb_steal_sock(skb, refcounted);
|
2008-10-07 18:41:57 +00:00
|
|
|
const struct iphdr *iph = ip_hdr(skb);
|
|
|
|
|
2012-06-20 04:22:05 +00:00
|
|
|
if (sk)
|
2008-10-07 19:41:01 +00:00
|
|
|
return sk;
|
2016-04-01 15:52:17 +00:00
|
|
|
|
|
|
|
return __inet_lookup(dev_net(skb_dst(skb)->dev), hashinfo, skb,
|
|
|
|
doff, iph->saddr, sport,
|
2017-08-07 15:44:17 +00:00
|
|
|
iph->daddr, dport, inet_iif(skb), sdif,
|
2016-04-01 15:52:17 +00:00
|
|
|
refcounted);
|
2008-10-07 18:41:57 +00:00
|
|
|
}
|
|
|
|
|
2015-03-18 21:05:35 +00:00
|
|
|
u32 inet6_ehashfn(const struct net *net,
|
|
|
|
const struct in6_addr *laddr, const u16 lport,
|
|
|
|
const struct in6_addr *faddr, const __be16 fport);
|
|
|
|
|
|
|
|
static inline void sk_daddr_set(struct sock *sk, __be32 addr)
|
|
|
|
{
|
|
|
|
sk->sk_daddr = addr; /* alias of inet_daddr */
|
|
|
|
#if IS_ENABLED(CONFIG_IPV6)
|
|
|
|
ipv6_addr_set_v4mapped(addr, &sk->sk_v6_daddr);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void sk_rcv_saddr_set(struct sock *sk, __be32 addr)
|
|
|
|
{
|
|
|
|
sk->sk_rcv_saddr = addr; /* alias of inet_rcv_saddr */
|
|
|
|
#if IS_ENABLED(CONFIG_IPV6)
|
|
|
|
ipv6_addr_set_v4mapped(addr, &sk->sk_v6_rcv_saddr);
|
|
|
|
#endif
|
|
|
|
}
|
2015-03-18 21:05:34 +00:00
|
|
|
|
2013-09-21 17:22:41 +00:00
|
|
|
int __inet_hash_connect(struct inet_timewait_death_row *death_row,
|
2022-05-02 08:46:08 +00:00
|
|
|
struct sock *sk, u64 port_offset,
|
2013-09-21 17:22:41 +00:00
|
|
|
int (*check_established)(struct inet_timewait_death_row *,
|
|
|
|
struct sock *, __u16,
|
2015-03-18 21:05:37 +00:00
|
|
|
struct inet_timewait_sock **));
|
2009-12-04 03:46:54 +00:00
|
|
|
|
2013-09-21 17:22:41 +00:00
|
|
|
int inet_hash_connect(struct inet_timewait_death_row *death_row,
|
|
|
|
struct sock *sk);
|
2005-08-10 02:59:20 +00:00
|
|
|
#endif /* _INET_HASHTABLES_H */
|