2005-04-17 06:20:36 +08:00
|
|
|
/*
|
|
|
|
* TCP over IPv6
|
2007-02-09 22:24:49 +08:00
|
|
|
* Linux INET6 implementation
|
2005-04-17 06:20:36 +08:00
|
|
|
*
|
|
|
|
* Authors:
|
2007-02-09 22:24:49 +08:00
|
|
|
* Pedro Roque <roque@di.fc.ul.pt>
|
2005-04-17 06:20:36 +08:00
|
|
|
*
|
2007-02-09 22:24:49 +08:00
|
|
|
* Based on:
|
2005-04-17 06:20:36 +08:00
|
|
|
* linux/net/ipv4/tcp.c
|
|
|
|
* linux/net/ipv4/tcp_input.c
|
|
|
|
* linux/net/ipv4/tcp_output.c
|
|
|
|
*
|
|
|
|
* Fixes:
|
|
|
|
* Hideaki YOSHIFUJI : sin6_scope_id support
|
|
|
|
* YOSHIFUJI Hideaki @USAGI and: Support IPV6_V6ONLY socket option, which
|
|
|
|
* Alexey Kuznetsov allow both IPv4 and IPv6 sockets to bind
|
|
|
|
* a single port at the same time.
|
|
|
|
* YOSHIFUJI Hideaki @USAGI: convert /proc/net/tcp6 to seq_file.
|
|
|
|
*
|
|
|
|
* 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.
|
|
|
|
*/
|
|
|
|
|
2008-12-30 15:04:08 +08:00
|
|
|
#include <linux/bottom_half.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/errno.h>
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/socket.h>
|
|
|
|
#include <linux/sockios.h>
|
|
|
|
#include <linux/net.h>
|
|
|
|
#include <linux/jiffies.h>
|
|
|
|
#include <linux/in.h>
|
|
|
|
#include <linux/in6.h>
|
|
|
|
#include <linux/netdevice.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/jhash.h>
|
|
|
|
#include <linux/ipsec.h>
|
|
|
|
#include <linux/times.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 16:04:11 +08:00
|
|
|
#include <linux/slab.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
#include <linux/ipv6.h>
|
|
|
|
#include <linux/icmpv6.h>
|
|
|
|
#include <linux/random.h>
|
|
|
|
|
|
|
|
#include <net/tcp.h>
|
|
|
|
#include <net/ndisc.h>
|
2005-08-12 20:26:18 +08:00
|
|
|
#include <net/inet6_hashtables.h>
|
2005-12-14 15:15:24 +08:00
|
|
|
#include <net/inet6_connection_sock.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
#include <net/ipv6.h>
|
|
|
|
#include <net/transp_v6.h>
|
|
|
|
#include <net/addrconf.h>
|
|
|
|
#include <net/ip6_route.h>
|
|
|
|
#include <net/ip6_checksum.h>
|
|
|
|
#include <net/inet_ecn.h>
|
|
|
|
#include <net/protocol.h>
|
|
|
|
#include <net/xfrm.h>
|
|
|
|
#include <net/snmp.h>
|
|
|
|
#include <net/dsfield.h>
|
2005-12-14 15:25:19 +08:00
|
|
|
#include <net/timewait_sock.h>
|
2007-10-27 13:53:14 +08:00
|
|
|
#include <net/netdma.h>
|
2008-04-04 05:22:32 +08:00
|
|
|
#include <net/inet_common.h>
|
2011-08-04 11:50:44 +08:00
|
|
|
#include <net/secure_seq.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
#include <asm/uaccess.h>
|
|
|
|
|
|
|
|
#include <linux/proc_fs.h>
|
|
|
|
#include <linux/seq_file.h>
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#include <linux/crypto.h>
|
|
|
|
#include <linux/scatterlist.h>
|
|
|
|
|
|
|
|
static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb);
|
2008-08-07 14:50:04 +08:00
|
|
|
static void tcp_v6_reqsk_send_ack(struct sock *sk, struct sk_buff *skb,
|
|
|
|
struct request_sock *req);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb);
|
2010-04-11 10:15:54 +08:00
|
|
|
static void __tcp_v6_send_check(struct sk_buff *skb,
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *saddr,
|
|
|
|
const struct in6_addr *daddr);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2009-09-02 03:25:04 +08:00
|
|
|
static const struct inet_connection_sock_af_ops ipv6_mapped;
|
|
|
|
static const struct inet_connection_sock_af_ops ipv6_specific;
|
2006-11-15 11:53:22 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
2009-09-02 03:25:03 +08:00
|
|
|
static const struct tcp_sock_af_ops tcp_sock_ipv6_specific;
|
|
|
|
static const struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific;
|
2008-04-18 11:45:16 +08:00
|
|
|
#else
|
|
|
|
static struct tcp_md5sig_key *tcp_v6_md5_do_lookup(struct sock *sk,
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *addr)
|
2008-04-18 11:45:16 +08:00
|
|
|
{
|
|
|
|
return NULL;
|
|
|
|
}
|
2006-11-15 11:53:22 +08:00
|
|
|
#endif
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
static void tcp_v6_hash(struct sock *sk)
|
|
|
|
{
|
|
|
|
if (sk->sk_state != TCP_CLOSE) {
|
2005-12-14 15:15:52 +08:00
|
|
|
if (inet_csk(sk)->icsk_af_ops == &ipv6_mapped) {
|
2005-04-17 06:20:36 +08:00
|
|
|
tcp_prot.hash(sk);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
local_bh_disable();
|
2009-12-04 11:46:54 +08:00
|
|
|
__inet6_hash(sk, NULL);
|
2005-04-17 06:20:36 +08:00
|
|
|
local_bh_enable();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-01-09 02:41:23 +08:00
|
|
|
static __inline__ __sum16 tcp_v6_check(int len,
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *saddr,
|
|
|
|
const struct in6_addr *daddr,
|
2006-11-15 13:35:48 +08:00
|
|
|
__wsum base)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
return csum_ipv6_magic(saddr, daddr, len, IPPROTO_TCP, base);
|
|
|
|
}
|
|
|
|
|
2011-10-21 17:22:42 +08:00
|
|
|
static __u32 tcp_v6_init_sequence(const struct sk_buff *skb)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2007-04-26 08:54:47 +08:00
|
|
|
return secure_tcpv6_sequence_number(ipv6_hdr(skb)->daddr.s6_addr32,
|
|
|
|
ipv6_hdr(skb)->saddr.s6_addr32,
|
2007-04-11 12:04:22 +08:00
|
|
|
tcp_hdr(skb)->dest,
|
|
|
|
tcp_hdr(skb)->source);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2007-02-09 22:24:49 +08:00
|
|
|
static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
|
2005-04-17 06:20:36 +08:00
|
|
|
int addr_len)
|
|
|
|
{
|
|
|
|
struct sockaddr_in6 *usin = (struct sockaddr_in6 *) uaddr;
|
2007-02-09 22:24:49 +08:00
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
2005-12-14 15:26:10 +08:00
|
|
|
struct inet_connection_sock *icsk = inet_csk(sk);
|
2005-04-17 06:20:36 +08:00
|
|
|
struct ipv6_pinfo *np = inet6_sk(sk);
|
|
|
|
struct tcp_sock *tp = tcp_sk(sk);
|
2010-06-02 05:35:01 +08:00
|
|
|
struct in6_addr *saddr = NULL, *final_p, final;
|
2010-12-03 04:14:29 +08:00
|
|
|
struct rt6_info *rt;
|
2011-03-13 05:22:43 +08:00
|
|
|
struct flowi6 fl6;
|
2005-04-17 06:20:36 +08:00
|
|
|
struct dst_entry *dst;
|
|
|
|
int addr_type;
|
|
|
|
int err;
|
|
|
|
|
2007-02-09 22:24:49 +08:00
|
|
|
if (addr_len < SIN6_LEN_RFC2133)
|
2005-04-17 06:20:36 +08:00
|
|
|
return -EINVAL;
|
|
|
|
|
2007-02-09 22:24:49 +08:00
|
|
|
if (usin->sin6_family != AF_INET6)
|
2010-09-23 04:43:57 +08:00
|
|
|
return -EAFNOSUPPORT;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
memset(&fl6, 0, sizeof(fl6));
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (np->sndflow) {
|
2011-03-13 05:22:43 +08:00
|
|
|
fl6.flowlabel = usin->sin6_flowinfo&IPV6_FLOWINFO_MASK;
|
|
|
|
IP6_ECN_flow_init(fl6.flowlabel);
|
|
|
|
if (fl6.flowlabel&IPV6_FLOWLABEL_MASK) {
|
2005-04-17 06:20:36 +08:00
|
|
|
struct ip6_flowlabel *flowlabel;
|
2011-03-13 05:22:43 +08:00
|
|
|
flowlabel = fl6_sock_lookup(sk, fl6.flowlabel);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (flowlabel == NULL)
|
|
|
|
return -EINVAL;
|
|
|
|
ipv6_addr_copy(&usin->sin6_addr, &flowlabel->dst);
|
|
|
|
fl6_sock_release(flowlabel);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-02-09 22:24:49 +08:00
|
|
|
* connect() to INADDR_ANY means loopback (BSD'ism).
|
|
|
|
*/
|
|
|
|
|
|
|
|
if(ipv6_addr_any(&usin->sin6_addr))
|
|
|
|
usin->sin6_addr.s6_addr[15] = 0x1;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
addr_type = ipv6_addr_type(&usin->sin6_addr);
|
|
|
|
|
|
|
|
if(addr_type & IPV6_ADDR_MULTICAST)
|
|
|
|
return -ENETUNREACH;
|
|
|
|
|
|
|
|
if (addr_type&IPV6_ADDR_LINKLOCAL) {
|
|
|
|
if (addr_len >= sizeof(struct sockaddr_in6) &&
|
|
|
|
usin->sin6_scope_id) {
|
|
|
|
/* If interface is set while binding, indices
|
|
|
|
* must coincide.
|
|
|
|
*/
|
|
|
|
if (sk->sk_bound_dev_if &&
|
|
|
|
sk->sk_bound_dev_if != usin->sin6_scope_id)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
sk->sk_bound_dev_if = usin->sin6_scope_id;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Connect to link-local address requires an interface */
|
|
|
|
if (!sk->sk_bound_dev_if)
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (tp->rx_opt.ts_recent_stamp &&
|
|
|
|
!ipv6_addr_equal(&np->daddr, &usin->sin6_addr)) {
|
|
|
|
tp->rx_opt.ts_recent = 0;
|
|
|
|
tp->rx_opt.ts_recent_stamp = 0;
|
|
|
|
tp->write_seq = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
ipv6_addr_copy(&np->daddr, &usin->sin6_addr);
|
2011-03-13 05:22:43 +08:00
|
|
|
np->flow_label = fl6.flowlabel;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* TCP over IPv4
|
|
|
|
*/
|
|
|
|
|
|
|
|
if (addr_type == IPV6_ADDR_MAPPED) {
|
2005-12-14 15:26:10 +08:00
|
|
|
u32 exthdrlen = icsk->icsk_ext_hdr_len;
|
2005-04-17 06:20:36 +08:00
|
|
|
struct sockaddr_in sin;
|
|
|
|
|
|
|
|
SOCK_DEBUG(sk, "connect: ipv4 mapped\n");
|
|
|
|
|
|
|
|
if (__ipv6_only_sock(sk))
|
|
|
|
return -ENETUNREACH;
|
|
|
|
|
|
|
|
sin.sin_family = AF_INET;
|
|
|
|
sin.sin_port = usin->sin6_port;
|
|
|
|
sin.sin_addr.s_addr = usin->sin6_addr.s6_addr32[3];
|
|
|
|
|
2005-12-14 15:26:10 +08:00
|
|
|
icsk->icsk_af_ops = &ipv6_mapped;
|
2005-04-17 06:20:36 +08:00
|
|
|
sk->sk_backlog_rcv = tcp_v4_do_rcv;
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
tp->af_specific = &tcp_sock_ipv6_mapped_specific;
|
|
|
|
#endif
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
err = tcp_v4_connect(sk, (struct sockaddr *)&sin, sizeof(sin));
|
|
|
|
|
|
|
|
if (err) {
|
2005-12-14 15:26:10 +08:00
|
|
|
icsk->icsk_ext_hdr_len = exthdrlen;
|
|
|
|
icsk->icsk_af_ops = &ipv6_specific;
|
2005-04-17 06:20:36 +08:00
|
|
|
sk->sk_backlog_rcv = tcp_v6_do_rcv;
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
tp->af_specific = &tcp_sock_ipv6_specific;
|
|
|
|
#endif
|
2005-04-17 06:20:36 +08:00
|
|
|
goto failure;
|
|
|
|
} else {
|
2009-10-15 14:30:45 +08:00
|
|
|
ipv6_addr_set_v4mapped(inet->inet_saddr, &np->saddr);
|
|
|
|
ipv6_addr_set_v4mapped(inet->inet_rcv_saddr,
|
|
|
|
&np->rcv_saddr);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!ipv6_addr_any(&np->rcv_saddr))
|
|
|
|
saddr = &np->rcv_saddr;
|
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
fl6.flowi6_proto = IPPROTO_TCP;
|
|
|
|
ipv6_addr_copy(&fl6.daddr, &np->daddr);
|
|
|
|
ipv6_addr_copy(&fl6.saddr,
|
2005-04-17 06:20:36 +08:00
|
|
|
(saddr ? saddr : &np->saddr));
|
2011-03-13 05:22:43 +08:00
|
|
|
fl6.flowi6_oif = sk->sk_bound_dev_if;
|
|
|
|
fl6.flowi6_mark = sk->sk_mark;
|
2011-03-13 05:36:19 +08:00
|
|
|
fl6.fl6_dport = usin->sin6_port;
|
|
|
|
fl6.fl6_sport = inet->inet_sport;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
final_p = fl6_update_dst(&fl6, np->opt, &final);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
security_sk_classify_flow(sk, flowi6_to_flowi(&fl6));
|
2006-08-05 14:12:42 +08:00
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
dst = ip6_dst_lookup_flow(sk, &fl6, final_p, true);
|
2011-03-02 05:19:07 +08:00
|
|
|
if (IS_ERR(dst)) {
|
|
|
|
err = PTR_ERR(dst);
|
2005-04-17 06:20:36 +08:00
|
|
|
goto failure;
|
2007-05-25 09:17:54 +08:00
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (saddr == NULL) {
|
2011-03-13 05:22:43 +08:00
|
|
|
saddr = &fl6.saddr;
|
2005-04-17 06:20:36 +08:00
|
|
|
ipv6_addr_copy(&np->rcv_saddr, saddr);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* set the source address */
|
|
|
|
ipv6_addr_copy(&np->saddr, saddr);
|
2009-10-15 14:30:45 +08:00
|
|
|
inet->inet_rcv_saddr = LOOPBACK4_IPV6;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-07-01 04:37:03 +08:00
|
|
|
sk->sk_gso_type = SKB_GSO_TCPV6;
|
2006-08-30 08:15:09 +08:00
|
|
|
__ip6_dst_store(sk, dst, NULL, NULL);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2010-12-03 04:14:29 +08:00
|
|
|
rt = (struct rt6_info *) dst;
|
|
|
|
if (tcp_death_row.sysctl_tw_recycle &&
|
|
|
|
!tp->rx_opt.ts_recent_stamp &&
|
|
|
|
ipv6_addr_equal(&rt->rt6i_dst.addr, &np->daddr)) {
|
|
|
|
struct inet_peer *peer = rt6_get_peer(rt);
|
|
|
|
/*
|
|
|
|
* VJ's idea. We save last timestamp seen from
|
|
|
|
* the destination in peer table, when entering state
|
|
|
|
* TIME-WAIT * and initialize rx_opt.ts_recent from it,
|
|
|
|
* when trying new connection.
|
|
|
|
*/
|
|
|
|
if (peer) {
|
|
|
|
inet_peer_refcheck(peer);
|
|
|
|
if ((u32)get_seconds() - peer->tcp_ts_stamp <= TCP_PAWS_MSL) {
|
|
|
|
tp->rx_opt.ts_recent_stamp = peer->tcp_ts_stamp;
|
|
|
|
tp->rx_opt.ts_recent = peer->tcp_ts;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2005-12-14 15:26:10 +08:00
|
|
|
icsk->icsk_ext_hdr_len = 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
if (np->opt)
|
2005-12-14 15:26:10 +08:00
|
|
|
icsk->icsk_ext_hdr_len = (np->opt->opt_flen +
|
|
|
|
np->opt->opt_nflen);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
tp->rx_opt.mss_clamp = IPV6_MIN_MTU - sizeof(struct tcphdr) - sizeof(struct ipv6hdr);
|
|
|
|
|
2009-10-15 14:30:45 +08:00
|
|
|
inet->inet_dport = usin->sin6_port;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
tcp_set_state(sk, TCP_SYN_SENT);
|
2005-12-14 15:25:44 +08:00
|
|
|
err = inet6_hash_connect(&tcp_death_row, sk);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (err)
|
|
|
|
goto late_failure;
|
|
|
|
|
|
|
|
if (!tp->write_seq)
|
|
|
|
tp->write_seq = secure_tcpv6_sequence_number(np->saddr.s6_addr32,
|
|
|
|
np->daddr.s6_addr32,
|
2009-10-15 14:30:45 +08:00
|
|
|
inet->inet_sport,
|
|
|
|
inet->inet_dport);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
err = tcp_connect(sk);
|
|
|
|
if (err)
|
|
|
|
goto late_failure;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
late_failure:
|
|
|
|
tcp_set_state(sk, TCP_CLOSE);
|
|
|
|
__sk_dst_reset(sk);
|
|
|
|
failure:
|
2009-10-15 14:30:45 +08:00
|
|
|
inet->inet_dport = 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
sk->sk_route_caps = 0;
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
|
2009-06-23 19:31:07 +08:00
|
|
|
u8 type, u8 code, int offset, __be32 info)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct ipv6hdr *hdr = (const struct ipv6hdr*)skb->data;
|
2005-08-12 20:19:38 +08:00
|
|
|
const struct tcphdr *th = (struct tcphdr *)(skb->data+offset);
|
2005-04-17 06:20:36 +08:00
|
|
|
struct ipv6_pinfo *np;
|
|
|
|
struct sock *sk;
|
|
|
|
int err;
|
2007-02-09 22:24:49 +08:00
|
|
|
struct tcp_sock *tp;
|
2005-04-17 06:20:36 +08:00
|
|
|
__u32 seq;
|
2008-07-17 11:28:42 +08:00
|
|
|
struct net *net = dev_net(skb->dev);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-07-17 11:28:42 +08:00
|
|
|
sk = inet6_lookup(net, &tcp_hashinfo, &hdr->daddr,
|
2008-01-31 21:07:21 +08:00
|
|
|
th->dest, &hdr->saddr, th->source, skb->dev->ifindex);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (sk == NULL) {
|
2008-10-09 01:33:26 +08:00
|
|
|
ICMP6_INC_STATS_BH(net, __in6_dev_get(skb->dev),
|
|
|
|
ICMP6_MIB_INERRORS);
|
2005-04-17 06:20:36 +08:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (sk->sk_state == TCP_TIME_WAIT) {
|
2006-10-11 10:41:46 +08:00
|
|
|
inet_twsk_put(inet_twsk(sk));
|
2005-04-17 06:20:36 +08:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
bh_lock_sock(sk);
|
|
|
|
if (sock_owned_by_user(sk))
|
2008-07-17 11:31:16 +08:00
|
|
|
NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (sk->sk_state == TCP_CLOSE)
|
|
|
|
goto out;
|
|
|
|
|
IPv6: Generic TTL Security Mechanism (final version)
This patch adds IPv6 support for RFC5082 Generalized TTL Security Mechanism.
Not to users of mapped address; the IPV6 and IPV4 socket options are seperate.
The server does have to deal with both IPv4 and IPv6 socket options
and the client has to handle the different for each family.
On client:
int ttl = 255;
getaddrinfo(argv[1], argv[2], &hint, &result);
for (rp = result; rp != NULL; rp = rp->ai_next) {
s = socket(rp->ai_family, rp->ai_socktype, rp->ai_protocol);
if (s < 0) continue;
if (rp->ai_family == AF_INET) {
setsockopt(s, IPPROTO_IP, IP_TTL, &ttl, sizeof(ttl));
} else if (rp->ai_family == AF_INET6) {
setsockopt(s, IPPROTO_IPV6, IPV6_UNICAST_HOPS,
&ttl, sizeof(ttl)))
}
if (connect(s, rp->ai_addr, rp->ai_addrlen) == 0) {
...
On server:
int minttl = 255 - maxhops;
getaddrinfo(NULL, port, &hints, &result);
for (rp = result; rp != NULL; rp = rp->ai_next) {
s = socket(rp->ai_family, rp->ai_socktype, rp->ai_protocol);
if (s < 0) continue;
if (rp->ai_family == AF_INET6)
setsockopt(s, IPPROTO_IPV6, IPV6_MINHOPCOUNT,
&minttl, sizeof(minttl));
setsockopt(s, IPPROTO_IP, IP_MINTTL, &minttl, sizeof(minttl));
if (bind(s, rp->ai_addr, rp->ai_addrlen) == 0)
break
...
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-23 06:24:53 +08:00
|
|
|
if (ipv6_hdr(skb)->hop_limit < inet6_sk(sk)->min_hopcount) {
|
|
|
|
NET_INC_STATS_BH(net, LINUX_MIB_TCPMINTTLDROP);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
tp = tcp_sk(sk);
|
2007-02-09 22:24:49 +08:00
|
|
|
seq = ntohl(th->seq);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (sk->sk_state != TCP_LISTEN &&
|
|
|
|
!between(seq, tp->snd_una, tp->snd_nxt)) {
|
2008-07-17 11:31:16 +08:00
|
|
|
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
|
2005-04-17 06:20:36 +08:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
np = inet6_sk(sk);
|
|
|
|
|
|
|
|
if (type == ICMPV6_PKT_TOOBIG) {
|
2011-03-02 05:19:07 +08:00
|
|
|
struct dst_entry *dst;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (sock_owned_by_user(sk))
|
|
|
|
goto out;
|
|
|
|
if ((1 << sk->sk_state) & (TCPF_LISTEN | TCPF_CLOSE))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
/* icmp should have updated the destination cache entry */
|
|
|
|
dst = __sk_dst_check(sk, np->dst_cookie);
|
|
|
|
|
|
|
|
if (dst == NULL) {
|
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
2011-03-13 05:22:43 +08:00
|
|
|
struct flowi6 fl6;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
/* BUGGG_FUTURE: Again, it is not clear how
|
|
|
|
to handle rthdr case. Ignore this complexity
|
|
|
|
for now.
|
|
|
|
*/
|
2011-03-13 05:22:43 +08:00
|
|
|
memset(&fl6, 0, sizeof(fl6));
|
|
|
|
fl6.flowi6_proto = IPPROTO_TCP;
|
|
|
|
ipv6_addr_copy(&fl6.daddr, &np->daddr);
|
|
|
|
ipv6_addr_copy(&fl6.saddr, &np->saddr);
|
|
|
|
fl6.flowi6_oif = sk->sk_bound_dev_if;
|
|
|
|
fl6.flowi6_mark = sk->sk_mark;
|
2011-03-13 05:36:19 +08:00
|
|
|
fl6.fl6_dport = inet->inet_dport;
|
|
|
|
fl6.fl6_sport = inet->inet_sport;
|
2011-03-13 05:22:43 +08:00
|
|
|
security_skb_classify_flow(skb, flowi6_to_flowi(&fl6));
|
|
|
|
|
|
|
|
dst = ip6_dst_lookup_flow(sk, &fl6, NULL, false);
|
2011-03-02 05:19:07 +08:00
|
|
|
if (IS_ERR(dst)) {
|
|
|
|
sk->sk_err_soft = -PTR_ERR(dst);
|
2005-04-17 06:20:36 +08:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
} else
|
|
|
|
dst_hold(dst);
|
|
|
|
|
2005-12-14 15:26:10 +08:00
|
|
|
if (inet_csk(sk)->icsk_pmtu_cookie > dst_mtu(dst)) {
|
2005-04-17 06:20:36 +08:00
|
|
|
tcp_sync_mss(sk, dst_mtu(dst));
|
|
|
|
tcp_simple_retransmit(sk);
|
|
|
|
} /* else let the usual retransmit timer handle it */
|
|
|
|
dst_release(dst);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
icmpv6_err_convert(type, code, &err);
|
|
|
|
|
2005-06-19 13:47:21 +08:00
|
|
|
/* Might be for an request_sock */
|
2005-04-17 06:20:36 +08:00
|
|
|
switch (sk->sk_state) {
|
2005-06-19 13:47:21 +08:00
|
|
|
struct request_sock *req, **prev;
|
2005-04-17 06:20:36 +08:00
|
|
|
case TCP_LISTEN:
|
|
|
|
if (sock_owned_by_user(sk))
|
|
|
|
goto out;
|
|
|
|
|
2005-12-14 15:15:24 +08:00
|
|
|
req = inet6_csk_search_req(sk, &prev, th->dest, &hdr->daddr,
|
|
|
|
&hdr->saddr, inet6_iif(skb));
|
2005-04-17 06:20:36 +08:00
|
|
|
if (!req)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
/* ICMPs are not backlogged, hence we cannot get
|
|
|
|
* an established socket here.
|
|
|
|
*/
|
2008-07-26 12:43:18 +08:00
|
|
|
WARN_ON(req->sk != NULL);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
[NET] Generalise TCP's struct open_request minisock infrastructure
Kept this first changeset minimal, without changing existing names to
ease peer review.
Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn
has two new members:
->slab, that replaces tcp_openreq_cachep
->obj_size, to inform the size of the openreq descendant for
a specific protocol
The protocol specific fields in struct open_request were moved to a
class hierarchy, with the things that are common to all connection
oriented PF_INET protocols in struct inet_request_sock, the TCP ones
in tcp_request_sock, that is an inet_request_sock, that is an
open_request.
I.e. this uses the same approach used for the struct sock class
hierarchy, with sk_prot indicating if the protocol wants to use the
open_request infrastructure by filling in sk_prot->rsk_prot with an
or_calltable.
Results? Performance is improved and TCP v4 now uses only 64 bytes per
open request minisock, down from 96 without this patch :-)
Next changeset will rename some of the structs, fields and functions
mentioned above, struct or_calltable is way unclear, better name it
struct request_sock_ops, s/struct open_request/struct request_sock/g,
etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-19 13:46:52 +08:00
|
|
|
if (seq != tcp_rsk(req)->snt_isn) {
|
2008-07-17 11:31:16 +08:00
|
|
|
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
|
2005-04-17 06:20:36 +08:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2005-08-10 11:10:42 +08:00
|
|
|
inet_csk_reqsk_queue_drop(sk, req, prev);
|
2005-04-17 06:20:36 +08:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
case TCP_SYN_SENT:
|
|
|
|
case TCP_SYN_RECV: /* Cannot happen.
|
2007-02-09 22:24:49 +08:00
|
|
|
It can, it SYNs are crossed. --ANK */
|
2005-04-17 06:20:36 +08:00
|
|
|
if (!sock_owned_by_user(sk)) {
|
|
|
|
sk->sk_err = err;
|
|
|
|
sk->sk_error_report(sk); /* Wake people up to see the error (see connect in sock.c) */
|
|
|
|
|
|
|
|
tcp_done(sk);
|
|
|
|
} else
|
|
|
|
sk->sk_err_soft = err;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!sock_owned_by_user(sk) && np->recverr) {
|
|
|
|
sk->sk_err = err;
|
|
|
|
sk->sk_error_report(sk);
|
|
|
|
} else
|
|
|
|
sk->sk_err_soft = err;
|
|
|
|
|
|
|
|
out:
|
|
|
|
bh_unlock_sock(sk);
|
|
|
|
sock_put(sk);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2009-12-03 02:07:39 +08:00
|
|
|
static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req,
|
|
|
|
struct request_values *rvp)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2005-12-14 15:15:40 +08:00
|
|
|
struct inet6_request_sock *treq = inet6_rsk(req);
|
2005-04-17 06:20:36 +08:00
|
|
|
struct ipv6_pinfo *np = inet6_sk(sk);
|
|
|
|
struct sk_buff * skb;
|
|
|
|
struct ipv6_txoptions *opt = NULL;
|
2010-06-02 05:35:01 +08:00
|
|
|
struct in6_addr * final_p, final;
|
2011-03-13 05:22:43 +08:00
|
|
|
struct flowi6 fl6;
|
2008-03-01 03:43:03 +08:00
|
|
|
struct dst_entry *dst;
|
2011-03-02 05:19:07 +08:00
|
|
|
int err;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
memset(&fl6, 0, sizeof(fl6));
|
|
|
|
fl6.flowi6_proto = IPPROTO_TCP;
|
|
|
|
ipv6_addr_copy(&fl6.daddr, &treq->rmt_addr);
|
|
|
|
ipv6_addr_copy(&fl6.saddr, &treq->loc_addr);
|
|
|
|
fl6.flowlabel = 0;
|
|
|
|
fl6.flowi6_oif = treq->iif;
|
|
|
|
fl6.flowi6_mark = sk->sk_mark;
|
2011-03-13 05:36:19 +08:00
|
|
|
fl6.fl6_dport = inet_rsk(req)->rmt_port;
|
|
|
|
fl6.fl6_sport = inet_rsk(req)->loc_port;
|
2011-03-13 05:22:43 +08:00
|
|
|
security_req_classify_flow(req, flowi6_to_flowi(&fl6));
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-03-01 03:43:03 +08:00
|
|
|
opt = np->opt;
|
2011-03-13 05:22:43 +08:00
|
|
|
final_p = fl6_update_dst(&fl6, opt, &final);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
dst = ip6_dst_lookup_flow(sk, &fl6, final_p, false);
|
2011-03-02 05:19:07 +08:00
|
|
|
if (IS_ERR(dst)) {
|
|
|
|
err = PTR_ERR(dst);
|
2011-04-05 04:07:26 +08:00
|
|
|
dst = NULL;
|
2008-03-01 03:43:03 +08:00
|
|
|
goto done;
|
2011-03-02 05:19:07 +08:00
|
|
|
}
|
2009-12-03 02:07:39 +08:00
|
|
|
skb = tcp_make_synack(sk, dst, req, rvp);
|
2011-03-02 05:19:07 +08:00
|
|
|
err = -ENOMEM;
|
2005-04-17 06:20:36 +08:00
|
|
|
if (skb) {
|
2010-04-11 10:15:54 +08:00
|
|
|
__tcp_v6_send_check(skb, &treq->loc_addr, &treq->rmt_addr);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
ipv6_addr_copy(&fl6.daddr, &treq->rmt_addr);
|
2011-10-27 12:44:35 +08:00
|
|
|
err = ip6_xmit(sk, skb, &fl6, opt, np->tclass);
|
2006-11-14 21:21:36 +08:00
|
|
|
err = net_xmit_eval(err);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
done:
|
2007-02-09 22:24:49 +08:00
|
|
|
if (opt && opt != np->opt)
|
2005-04-17 06:20:36 +08:00
|
|
|
sock_kfree_s(sk, opt, opt->tot_len);
|
2006-02-01 09:51:44 +08:00
|
|
|
dst_release(dst);
|
2005-04-17 06:20:36 +08:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2010-01-18 11:09:39 +08:00
|
|
|
static int tcp_v6_rtx_synack(struct sock *sk, struct request_sock *req,
|
|
|
|
struct request_values *rvp)
|
|
|
|
{
|
|
|
|
TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_RETRANSSEGS);
|
|
|
|
return tcp_v6_send_synack(sk, req, rvp);
|
|
|
|
}
|
|
|
|
|
2005-06-19 13:47:21 +08:00
|
|
|
static void tcp_v6_reqsk_destructor(struct request_sock *req)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2009-02-24 05:45:33 +08:00
|
|
|
kfree_skb(inet6_rsk(req)->pktopts);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
static struct tcp_md5sig_key *tcp_v6_md5_do_lookup(struct sock *sk,
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *addr)
|
2006-11-15 11:07:45 +08:00
|
|
|
{
|
|
|
|
struct tcp_sock *tp = tcp_sk(sk);
|
|
|
|
int i;
|
|
|
|
|
|
|
|
BUG_ON(tp == NULL);
|
|
|
|
|
|
|
|
if (!tp->md5sig_info || !tp->md5sig_info->entries6)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
for (i = 0; i < tp->md5sig_info->entries6; i++) {
|
2008-04-10 14:42:07 +08:00
|
|
|
if (ipv6_addr_equal(&tp->md5sig_info->keys6[i].addr, addr))
|
2007-09-29 06:18:35 +08:00
|
|
|
return &tp->md5sig_info->keys6[i].base;
|
2006-11-15 11:07:45 +08:00
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct tcp_md5sig_key *tcp_v6_md5_lookup(struct sock *sk,
|
|
|
|
struct sock *addr_sk)
|
|
|
|
{
|
|
|
|
return tcp_v6_md5_do_lookup(sk, &inet6_sk(addr_sk)->daddr);
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct tcp_md5sig_key *tcp_v6_reqsk_md5_lookup(struct sock *sk,
|
|
|
|
struct request_sock *req)
|
|
|
|
{
|
|
|
|
return tcp_v6_md5_do_lookup(sk, &inet6_rsk(req)->rmt_addr);
|
|
|
|
}
|
|
|
|
|
2011-04-22 12:53:02 +08:00
|
|
|
static int tcp_v6_md5_do_add(struct sock *sk, const struct in6_addr *peer,
|
2006-11-15 11:07:45 +08:00
|
|
|
char *newkey, u8 newkeylen)
|
|
|
|
{
|
|
|
|
/* Add key to the list */
|
2007-10-30 11:55:27 +08:00
|
|
|
struct tcp_md5sig_key *key;
|
2006-11-15 11:07:45 +08:00
|
|
|
struct tcp_sock *tp = tcp_sk(sk);
|
|
|
|
struct tcp6_md5sig_key *keys;
|
|
|
|
|
2007-10-30 11:55:27 +08:00
|
|
|
key = tcp_v6_md5_do_lookup(sk, peer);
|
2006-11-15 11:07:45 +08:00
|
|
|
if (key) {
|
|
|
|
/* modify existing entry - just update that one */
|
2007-10-30 11:55:27 +08:00
|
|
|
kfree(key->key);
|
|
|
|
key->key = newkey;
|
|
|
|
key->keylen = newkeylen;
|
2006-11-15 11:07:45 +08:00
|
|
|
} else {
|
|
|
|
/* reallocate new list if current one is full. */
|
|
|
|
if (!tp->md5sig_info) {
|
|
|
|
tp->md5sig_info = kzalloc(sizeof(*tp->md5sig_info), GFP_ATOMIC);
|
|
|
|
if (!tp->md5sig_info) {
|
|
|
|
kfree(newkey);
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
2010-05-16 15:36:33 +08:00
|
|
|
sk_nocaps_add(sk, NETIF_F_GSO_MASK);
|
2006-11-15 11:07:45 +08:00
|
|
|
}
|
2011-09-30 01:10:10 +08:00
|
|
|
if (tp->md5sig_info->entries6 == 0 &&
|
|
|
|
tcp_alloc_md5sig_pool(sk) == NULL) {
|
2007-11-21 09:30:56 +08:00
|
|
|
kfree(newkey);
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
2006-11-15 11:07:45 +08:00
|
|
|
if (tp->md5sig_info->alloced6 == tp->md5sig_info->entries6) {
|
|
|
|
keys = kmalloc((sizeof (tp->md5sig_info->keys6[0]) *
|
|
|
|
(tp->md5sig_info->entries6 + 1)), GFP_ATOMIC);
|
|
|
|
|
|
|
|
if (!keys) {
|
|
|
|
kfree(newkey);
|
2011-09-30 01:10:10 +08:00
|
|
|
if (tp->md5sig_info->entries6 == 0)
|
|
|
|
tcp_free_md5sig_pool();
|
2006-11-15 11:07:45 +08:00
|
|
|
return -ENOMEM;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (tp->md5sig_info->entries6)
|
|
|
|
memmove(keys, tp->md5sig_info->keys6,
|
|
|
|
(sizeof (tp->md5sig_info->keys6[0]) *
|
|
|
|
tp->md5sig_info->entries6));
|
|
|
|
|
|
|
|
kfree(tp->md5sig_info->keys6);
|
|
|
|
tp->md5sig_info->keys6 = keys;
|
|
|
|
tp->md5sig_info->alloced6++;
|
|
|
|
}
|
|
|
|
|
|
|
|
ipv6_addr_copy(&tp->md5sig_info->keys6[tp->md5sig_info->entries6].addr,
|
|
|
|
peer);
|
2007-09-29 06:18:35 +08:00
|
|
|
tp->md5sig_info->keys6[tp->md5sig_info->entries6].base.key = newkey;
|
|
|
|
tp->md5sig_info->keys6[tp->md5sig_info->entries6].base.keylen = newkeylen;
|
2006-11-15 11:07:45 +08:00
|
|
|
|
|
|
|
tp->md5sig_info->entries6++;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int tcp_v6_md5_add_func(struct sock *sk, struct sock *addr_sk,
|
|
|
|
u8 *newkey, __u8 newkeylen)
|
|
|
|
{
|
|
|
|
return tcp_v6_md5_do_add(sk, &inet6_sk(addr_sk)->daddr,
|
|
|
|
newkey, newkeylen);
|
|
|
|
}
|
|
|
|
|
2011-04-22 12:53:02 +08:00
|
|
|
static int tcp_v6_md5_do_del(struct sock *sk, const struct in6_addr *peer)
|
2006-11-15 11:07:45 +08:00
|
|
|
{
|
|
|
|
struct tcp_sock *tp = tcp_sk(sk);
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < tp->md5sig_info->entries6; i++) {
|
2008-04-10 14:42:07 +08:00
|
|
|
if (ipv6_addr_equal(&tp->md5sig_info->keys6[i].addr, peer)) {
|
2006-11-15 11:07:45 +08:00
|
|
|
/* Free the key */
|
2007-09-29 06:18:35 +08:00
|
|
|
kfree(tp->md5sig_info->keys6[i].base.key);
|
2006-11-15 11:07:45 +08:00
|
|
|
tp->md5sig_info->entries6--;
|
|
|
|
|
|
|
|
if (tp->md5sig_info->entries6 == 0) {
|
|
|
|
kfree(tp->md5sig_info->keys6);
|
|
|
|
tp->md5sig_info->keys6 = NULL;
|
2007-07-25 06:27:30 +08:00
|
|
|
tp->md5sig_info->alloced6 = 0;
|
2011-09-30 01:10:10 +08:00
|
|
|
tcp_free_md5sig_pool();
|
2006-11-15 11:07:45 +08:00
|
|
|
} else {
|
|
|
|
/* shrink the database */
|
|
|
|
if (tp->md5sig_info->entries6 != i)
|
|
|
|
memmove(&tp->md5sig_info->keys6[i],
|
|
|
|
&tp->md5sig_info->keys6[i+1],
|
|
|
|
(tp->md5sig_info->entries6 - i)
|
|
|
|
* sizeof (tp->md5sig_info->keys6[0]));
|
|
|
|
}
|
2007-11-21 09:31:23 +08:00
|
|
|
return 0;
|
2006-11-15 11:07:45 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
return -ENOENT;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void tcp_v6_clear_md5_list (struct sock *sk)
|
|
|
|
{
|
|
|
|
struct tcp_sock *tp = tcp_sk(sk);
|
|
|
|
int i;
|
|
|
|
|
|
|
|
if (tp->md5sig_info->entries6) {
|
|
|
|
for (i = 0; i < tp->md5sig_info->entries6; i++)
|
2007-09-29 06:18:35 +08:00
|
|
|
kfree(tp->md5sig_info->keys6[i].base.key);
|
2006-11-15 11:07:45 +08:00
|
|
|
tp->md5sig_info->entries6 = 0;
|
|
|
|
tcp_free_md5sig_pool();
|
|
|
|
}
|
|
|
|
|
|
|
|
kfree(tp->md5sig_info->keys6);
|
|
|
|
tp->md5sig_info->keys6 = NULL;
|
|
|
|
tp->md5sig_info->alloced6 = 0;
|
|
|
|
|
|
|
|
if (tp->md5sig_info->entries4) {
|
|
|
|
for (i = 0; i < tp->md5sig_info->entries4; i++)
|
2007-09-29 06:18:35 +08:00
|
|
|
kfree(tp->md5sig_info->keys4[i].base.key);
|
2006-11-15 11:07:45 +08:00
|
|
|
tp->md5sig_info->entries4 = 0;
|
|
|
|
tcp_free_md5sig_pool();
|
|
|
|
}
|
|
|
|
|
|
|
|
kfree(tp->md5sig_info->keys4);
|
|
|
|
tp->md5sig_info->keys4 = NULL;
|
|
|
|
tp->md5sig_info->alloced4 = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int tcp_v6_parse_md5_keys (struct sock *sk, char __user *optval,
|
|
|
|
int optlen)
|
|
|
|
{
|
|
|
|
struct tcp_md5sig cmd;
|
|
|
|
struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&cmd.tcpm_addr;
|
|
|
|
u8 *newkey;
|
|
|
|
|
|
|
|
if (optlen < sizeof(cmd))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (copy_from_user(&cmd, optval, sizeof(cmd)))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
if (sin6->sin6_family != AF_INET6)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (!cmd.tcpm_keylen) {
|
|
|
|
if (!tcp_sk(sk)->md5sig_info)
|
|
|
|
return -ENOENT;
|
2007-08-25 14:16:08 +08:00
|
|
|
if (ipv6_addr_v4mapped(&sin6->sin6_addr))
|
2006-11-15 11:07:45 +08:00
|
|
|
return tcp_v4_md5_do_del(sk, sin6->sin6_addr.s6_addr32[3]);
|
|
|
|
return tcp_v6_md5_do_del(sk, &sin6->sin6_addr);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (cmd.tcpm_keylen > TCP_MD5SIG_MAXKEYLEN)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (!tcp_sk(sk)->md5sig_info) {
|
|
|
|
struct tcp_sock *tp = tcp_sk(sk);
|
|
|
|
struct tcp_md5sig_info *p;
|
|
|
|
|
|
|
|
p = kzalloc(sizeof(struct tcp_md5sig_info), GFP_KERNEL);
|
|
|
|
if (!p)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
tp->md5sig_info = p;
|
2010-05-16 15:36:33 +08:00
|
|
|
sk_nocaps_add(sk, NETIF_F_GSO_MASK);
|
2006-11-15 11:07:45 +08:00
|
|
|
}
|
|
|
|
|
2006-11-17 22:14:37 +08:00
|
|
|
newkey = kmemdup(cmd.tcpm_key, cmd.tcpm_keylen, GFP_KERNEL);
|
2006-11-15 11:07:45 +08:00
|
|
|
if (!newkey)
|
|
|
|
return -ENOMEM;
|
2007-08-25 14:16:08 +08:00
|
|
|
if (ipv6_addr_v4mapped(&sin6->sin6_addr)) {
|
2006-11-15 11:07:45 +08:00
|
|
|
return tcp_v4_md5_do_add(sk, sin6->sin6_addr.s6_addr32[3],
|
|
|
|
newkey, cmd.tcpm_keylen);
|
|
|
|
}
|
|
|
|
return tcp_v6_md5_do_add(sk, &sin6->sin6_addr, newkey, cmd.tcpm_keylen);
|
|
|
|
}
|
|
|
|
|
2008-07-19 15:01:42 +08:00
|
|
|
static int tcp_v6_md5_hash_pseudoheader(struct tcp_md5sig_pool *hp,
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *daddr,
|
|
|
|
const struct in6_addr *saddr, int nbytes)
|
2006-11-15 11:07:45 +08:00
|
|
|
{
|
|
|
|
struct tcp6_pseudohdr *bp;
|
2008-07-19 15:01:42 +08:00
|
|
|
struct scatterlist sg;
|
2008-04-17 12:19:16 +08:00
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
bp = &hp->md5_blk.ip6;
|
|
|
|
/* 1. TCP pseudo-header (RFC2460) */
|
|
|
|
ipv6_addr_copy(&bp->saddr, saddr);
|
|
|
|
ipv6_addr_copy(&bp->daddr, daddr);
|
2008-07-19 15:01:42 +08:00
|
|
|
bp->protocol = cpu_to_be32(IPPROTO_TCP);
|
2008-08-01 12:36:07 +08:00
|
|
|
bp->len = cpu_to_be32(nbytes);
|
2006-11-15 11:07:45 +08:00
|
|
|
|
2008-07-19 15:01:42 +08:00
|
|
|
sg_init_one(&sg, bp, sizeof(*bp));
|
|
|
|
return crypto_hash_update(&hp->md5_desc, &sg, sizeof(*bp));
|
|
|
|
}
|
2007-10-26 15:41:21 +08:00
|
|
|
|
2008-07-19 15:01:42 +08:00
|
|
|
static int tcp_v6_md5_hash_hdr(char *md5_hash, struct tcp_md5sig_key *key,
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *daddr, struct in6_addr *saddr,
|
2011-10-24 14:46:04 +08:00
|
|
|
const struct tcphdr *th)
|
2008-07-19 15:01:42 +08:00
|
|
|
{
|
|
|
|
struct tcp_md5sig_pool *hp;
|
|
|
|
struct hash_desc *desc;
|
|
|
|
|
|
|
|
hp = tcp_get_md5sig_pool();
|
|
|
|
if (!hp)
|
|
|
|
goto clear_hash_noput;
|
|
|
|
desc = &hp->md5_desc;
|
|
|
|
|
|
|
|
if (crypto_hash_init(desc))
|
|
|
|
goto clear_hash;
|
|
|
|
if (tcp_v6_md5_hash_pseudoheader(hp, daddr, saddr, th->doff << 2))
|
|
|
|
goto clear_hash;
|
|
|
|
if (tcp_md5_hash_header(hp, th))
|
|
|
|
goto clear_hash;
|
|
|
|
if (tcp_md5_hash_key(hp, key))
|
|
|
|
goto clear_hash;
|
|
|
|
if (crypto_hash_final(desc, md5_hash))
|
2006-11-15 11:07:45 +08:00
|
|
|
goto clear_hash;
|
|
|
|
|
|
|
|
tcp_put_md5sig_pool();
|
|
|
|
return 0;
|
2008-07-19 15:01:42 +08:00
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
clear_hash:
|
|
|
|
tcp_put_md5sig_pool();
|
|
|
|
clear_hash_noput:
|
|
|
|
memset(md5_hash, 0, 16);
|
2008-07-19 15:01:42 +08:00
|
|
|
return 1;
|
2006-11-15 11:07:45 +08:00
|
|
|
}
|
|
|
|
|
2008-07-19 15:01:42 +08:00
|
|
|
static int tcp_v6_md5_hash_skb(char *md5_hash, struct tcp_md5sig_key *key,
|
2011-10-24 14:46:04 +08:00
|
|
|
const struct sock *sk,
|
|
|
|
const struct request_sock *req,
|
|
|
|
const struct sk_buff *skb)
|
2006-11-15 11:07:45 +08:00
|
|
|
{
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *saddr, *daddr;
|
2008-07-19 15:01:42 +08:00
|
|
|
struct tcp_md5sig_pool *hp;
|
|
|
|
struct hash_desc *desc;
|
2011-10-24 14:46:04 +08:00
|
|
|
const struct tcphdr *th = tcp_hdr(skb);
|
2006-11-15 11:07:45 +08:00
|
|
|
|
|
|
|
if (sk) {
|
|
|
|
saddr = &inet6_sk(sk)->saddr;
|
|
|
|
daddr = &inet6_sk(sk)->daddr;
|
2008-07-19 15:01:42 +08:00
|
|
|
} else if (req) {
|
2006-11-15 11:07:45 +08:00
|
|
|
saddr = &inet6_rsk(req)->loc_addr;
|
|
|
|
daddr = &inet6_rsk(req)->rmt_addr;
|
2008-07-19 15:01:42 +08:00
|
|
|
} else {
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct ipv6hdr *ip6h = ipv6_hdr(skb);
|
2008-07-19 15:01:42 +08:00
|
|
|
saddr = &ip6h->saddr;
|
|
|
|
daddr = &ip6h->daddr;
|
2006-11-15 11:07:45 +08:00
|
|
|
}
|
2008-07-19 15:01:42 +08:00
|
|
|
|
|
|
|
hp = tcp_get_md5sig_pool();
|
|
|
|
if (!hp)
|
|
|
|
goto clear_hash_noput;
|
|
|
|
desc = &hp->md5_desc;
|
|
|
|
|
|
|
|
if (crypto_hash_init(desc))
|
|
|
|
goto clear_hash;
|
|
|
|
|
|
|
|
if (tcp_v6_md5_hash_pseudoheader(hp, daddr, saddr, skb->len))
|
|
|
|
goto clear_hash;
|
|
|
|
if (tcp_md5_hash_header(hp, th))
|
|
|
|
goto clear_hash;
|
|
|
|
if (tcp_md5_hash_skb_data(hp, skb, th->doff << 2))
|
|
|
|
goto clear_hash;
|
|
|
|
if (tcp_md5_hash_key(hp, key))
|
|
|
|
goto clear_hash;
|
|
|
|
if (crypto_hash_final(desc, md5_hash))
|
|
|
|
goto clear_hash;
|
|
|
|
|
|
|
|
tcp_put_md5sig_pool();
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
clear_hash:
|
|
|
|
tcp_put_md5sig_pool();
|
|
|
|
clear_hash_noput:
|
|
|
|
memset(md5_hash, 0, 16);
|
|
|
|
return 1;
|
2006-11-15 11:07:45 +08:00
|
|
|
}
|
|
|
|
|
2011-10-24 14:46:04 +08:00
|
|
|
static int tcp_v6_inbound_md5_hash(struct sock *sk, const struct sk_buff *skb)
|
2006-11-15 11:07:45 +08:00
|
|
|
{
|
2011-10-21 17:22:42 +08:00
|
|
|
const __u8 *hash_location = NULL;
|
2006-11-15 11:07:45 +08:00
|
|
|
struct tcp_md5sig_key *hash_expected;
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct ipv6hdr *ip6h = ipv6_hdr(skb);
|
2011-10-24 14:46:04 +08:00
|
|
|
const struct tcphdr *th = tcp_hdr(skb);
|
2006-11-15 11:07:45 +08:00
|
|
|
int genhash;
|
|
|
|
u8 newhash[16];
|
|
|
|
|
|
|
|
hash_expected = tcp_v6_md5_do_lookup(sk, &ip6h->saddr);
|
2008-04-17 11:29:53 +08:00
|
|
|
hash_location = tcp_parse_md5sig_option(th);
|
2006-11-15 11:07:45 +08:00
|
|
|
|
2008-07-30 18:03:15 +08:00
|
|
|
/* We've parsed the options - do we have a hash? */
|
|
|
|
if (!hash_expected && !hash_location)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (hash_expected && !hash_location) {
|
|
|
|
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPMD5NOTFOUND);
|
2006-11-15 11:07:45 +08:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2008-07-30 18:03:15 +08:00
|
|
|
if (!hash_expected && hash_location) {
|
|
|
|
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPMD5UNEXPECTED);
|
2006-11-15 11:07:45 +08:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* check the signature */
|
2008-07-19 15:01:42 +08:00
|
|
|
genhash = tcp_v6_md5_hash_skb(newhash,
|
|
|
|
hash_expected,
|
|
|
|
NULL, NULL, skb);
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
if (genhash || memcmp(hash_location, newhash, 16) != 0) {
|
|
|
|
if (net_ratelimit()) {
|
2010-01-08 16:59:52 +08:00
|
|
|
printk(KERN_INFO "MD5 Hash %s for [%pI6c]:%u->[%pI6c]:%u\n",
|
2006-11-15 11:07:45 +08:00
|
|
|
genhash ? "failed" : "mismatch",
|
2008-10-29 07:09:23 +08:00
|
|
|
&ip6h->saddr, ntohs(th->source),
|
|
|
|
&ip6h->daddr, ntohs(th->dest));
|
2006-11-15 11:07:45 +08:00
|
|
|
}
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2008-02-08 13:49:26 +08:00
|
|
|
struct request_sock_ops tcp6_request_sock_ops __read_mostly = {
|
2005-04-17 06:20:36 +08:00
|
|
|
.family = AF_INET6,
|
[NET] Generalise TCP's struct open_request minisock infrastructure
Kept this first changeset minimal, without changing existing names to
ease peer review.
Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn
has two new members:
->slab, that replaces tcp_openreq_cachep
->obj_size, to inform the size of the openreq descendant for
a specific protocol
The protocol specific fields in struct open_request were moved to a
class hierarchy, with the things that are common to all connection
oriented PF_INET protocols in struct inet_request_sock, the TCP ones
in tcp_request_sock, that is an inet_request_sock, that is an
open_request.
I.e. this uses the same approach used for the struct sock class
hierarchy, with sk_prot indicating if the protocol wants to use the
open_request infrastructure by filling in sk_prot->rsk_prot with an
or_calltable.
Results? Performance is improved and TCP v4 now uses only 64 bytes per
open request minisock, down from 96 without this patch :-)
Next changeset will rename some of the structs, fields and functions
mentioned above, struct or_calltable is way unclear, better name it
struct request_sock_ops, s/struct open_request/struct request_sock/g,
etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-19 13:46:52 +08:00
|
|
|
.obj_size = sizeof(struct tcp6_request_sock),
|
2010-01-18 11:09:39 +08:00
|
|
|
.rtx_syn_ack = tcp_v6_rtx_synack,
|
2005-06-19 13:47:21 +08:00
|
|
|
.send_ack = tcp_v6_reqsk_send_ack,
|
|
|
|
.destructor = tcp_v6_reqsk_destructor,
|
2010-01-18 11:09:39 +08:00
|
|
|
.send_reset = tcp_v6_send_reset,
|
|
|
|
.syn_ack_timeout = tcp_syn_ack_timeout,
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
2009-09-02 03:25:03 +08:00
|
|
|
static const struct tcp_request_sock_ops tcp_request_sock_ipv6_ops = {
|
2006-11-15 11:07:45 +08:00
|
|
|
.md5_lookup = tcp_v6_reqsk_md5_lookup,
|
2009-07-16 13:04:51 +08:00
|
|
|
.calc_md5_hash = tcp_v6_md5_hash_skb,
|
2006-11-15 11:07:45 +08:00
|
|
|
};
|
2006-12-01 11:16:28 +08:00
|
|
|
#endif
|
2006-11-15 11:07:45 +08:00
|
|
|
|
2010-04-11 10:15:54 +08:00
|
|
|
static void __tcp_v6_send_check(struct sk_buff *skb,
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *saddr, const struct in6_addr *daddr)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2007-04-11 12:04:22 +08:00
|
|
|
struct tcphdr *th = tcp_hdr(skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-08-30 07:44:56 +08:00
|
|
|
if (skb->ip_summed == CHECKSUM_PARTIAL) {
|
2010-04-11 10:15:54 +08:00
|
|
|
th->check = ~tcp_v6_check(skb->len, saddr, daddr, 0);
|
2007-04-10 02:59:07 +08:00
|
|
|
skb->csum_start = skb_transport_header(skb) - skb->head;
|
2006-11-21 10:07:29 +08:00
|
|
|
skb->csum_offset = offsetof(struct tcphdr, check);
|
2005-04-17 06:20:36 +08:00
|
|
|
} else {
|
2010-04-11 10:15:54 +08:00
|
|
|
th->check = tcp_v6_check(skb->len, saddr, daddr,
|
|
|
|
csum_partial(th, th->doff << 2,
|
|
|
|
skb->csum));
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-04-11 10:15:55 +08:00
|
|
|
static void tcp_v6_send_check(struct sock *sk, struct sk_buff *skb)
|
2010-04-11 10:15:54 +08:00
|
|
|
{
|
|
|
|
struct ipv6_pinfo *np = inet6_sk(sk);
|
|
|
|
|
|
|
|
__tcp_v6_send_check(skb, &np->saddr, &np->daddr);
|
|
|
|
}
|
|
|
|
|
2006-07-09 04:34:56 +08:00
|
|
|
static int tcp_v6_gso_send_check(struct sk_buff *skb)
|
|
|
|
{
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct ipv6hdr *ipv6h;
|
2006-07-09 04:34:56 +08:00
|
|
|
struct tcphdr *th;
|
|
|
|
|
|
|
|
if (!pskb_may_pull(skb, sizeof(*th)))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2007-04-26 08:54:47 +08:00
|
|
|
ipv6h = ipv6_hdr(skb);
|
2007-04-11 12:04:22 +08:00
|
|
|
th = tcp_hdr(skb);
|
2006-07-09 04:34:56 +08:00
|
|
|
|
|
|
|
th->check = 0;
|
2006-08-30 07:44:56 +08:00
|
|
|
skb->ip_summed = CHECKSUM_PARTIAL;
|
2010-04-11 10:15:54 +08:00
|
|
|
__tcp_v6_send_check(skb, &ipv6h->saddr, &ipv6h->daddr);
|
2006-07-09 04:34:56 +08:00
|
|
|
return 0;
|
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2009-05-22 15:45:28 +08:00
|
|
|
static struct sk_buff **tcp6_gro_receive(struct sk_buff **head,
|
|
|
|
struct sk_buff *skb)
|
2009-01-09 02:41:23 +08:00
|
|
|
{
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct ipv6hdr *iph = skb_gro_network_header(skb);
|
2009-01-09 02:41:23 +08:00
|
|
|
|
|
|
|
switch (skb->ip_summed) {
|
|
|
|
case CHECKSUM_COMPLETE:
|
2009-01-29 22:19:50 +08:00
|
|
|
if (!tcp_v6_check(skb_gro_len(skb), &iph->saddr, &iph->daddr,
|
2009-01-09 02:41:23 +08:00
|
|
|
skb->csum)) {
|
|
|
|
skb->ip_summed = CHECKSUM_UNNECESSARY;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* fall through */
|
|
|
|
case CHECKSUM_NONE:
|
|
|
|
NAPI_GRO_CB(skb)->flush = 1;
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
return tcp_gro_receive(head, skb);
|
|
|
|
}
|
|
|
|
|
2009-05-22 15:45:28 +08:00
|
|
|
static int tcp6_gro_complete(struct sk_buff *skb)
|
2009-01-09 02:41:23 +08:00
|
|
|
{
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct ipv6hdr *iph = ipv6_hdr(skb);
|
2009-01-09 02:41:23 +08:00
|
|
|
struct tcphdr *th = tcp_hdr(skb);
|
|
|
|
|
|
|
|
th->check = ~tcp_v6_check(skb->len - skb_transport_offset(skb),
|
|
|
|
&iph->saddr, &iph->daddr, 0);
|
|
|
|
skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6;
|
|
|
|
|
|
|
|
return tcp_gro_complete(skb);
|
|
|
|
}
|
|
|
|
|
2008-10-10 05:42:40 +08:00
|
|
|
static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win,
|
2011-10-27 12:44:35 +08:00
|
|
|
u32 ts, struct tcp_md5sig_key *key, int rst, u8 tclass)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2011-10-21 17:22:42 +08:00
|
|
|
const struct tcphdr *th = tcp_hdr(skb);
|
|
|
|
struct tcphdr *t1;
|
2005-04-17 06:20:36 +08:00
|
|
|
struct sk_buff *buff;
|
2011-03-13 05:22:43 +08:00
|
|
|
struct flowi6 fl6;
|
2009-06-02 13:19:30 +08:00
|
|
|
struct net *net = dev_net(skb_dst(skb)->dev);
|
2008-03-08 03:16:26 +08:00
|
|
|
struct sock *ctl_sk = net->ipv6.tcp_sk;
|
2008-10-10 05:41:38 +08:00
|
|
|
unsigned int tot_len = sizeof(struct tcphdr);
|
2009-06-02 13:19:30 +08:00
|
|
|
struct dst_entry *dst;
|
tcpv6: convert opt[] -> topt in tcp_v6_send_reset
after this I get:
$ diff-funcs tcp_v6_send_reset tcp_ipv6.c tcp_ipv6.c tcp_v6_send_ack
--- tcp_ipv6.c:tcp_v6_send_reset()
+++ tcp_ipv6.c:tcp_v6_send_ack()
@@ -1,4 +1,5 @@
-static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
+static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win,
u32 ts,
+ struct tcp_md5sig_key *key)
{
struct tcphdr *th = tcp_hdr(skb), *t1;
struct sk_buff *buff;
@@ -7,31 +8,14 @@
struct sock *ctl_sk = net->ipv6.tcp_sk;
unsigned int tot_len = sizeof(struct tcphdr);
__be32 *topt;
-#ifdef CONFIG_TCP_MD5SIG
- struct tcp_md5sig_key *key;
-#endif
-
- if (th->rst)
- return;
-
- if (!ipv6_unicast_destination(skb))
- return;
+ if (ts)
+ tot_len += TCPOLEN_TSTAMP_ALIGNED;
#ifdef CONFIG_TCP_MD5SIG
- if (sk)
- key = tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->daddr);
- else
- key = NULL;
-
if (key)
tot_len += TCPOLEN_MD5SIG_ALIGNED;
#endif
- /*
- * We need to grab some memory, and put together an RST,
- * and then put it into the queue to be sent.
- */
-
buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len,
GFP_ATOMIC);
if (buff == NULL)
@@ -46,18 +30,20 @@
t1->dest = th->source;
t1->source = th->dest;
t1->doff = tot_len / 4;
- t1->rst = 1;
-
- if(th->ack) {
- t1->seq = th->ack_seq;
- } else {
- t1->ack = 1;
- t1->ack_seq = htonl(ntohl(th->seq) + th->syn + th->fin
- + skb->len - (th->doff<<2));
- }
+ t1->seq = htonl(seq);
+ t1->ack_seq = htonl(ack);
+ t1->ack = 1;
+ t1->window = htons(win);
topt = (__be32 *)(t1 + 1);
+ if (ts) {
+ *topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
+ (TCPOPT_TIMESTAMP << 8) |
TCPOLEN_TIMESTAMP);
+ *topt++ = htonl(tcp_time_stamp);
+ *topt++ = htonl(ts);
+ }
+
#ifdef CONFIG_TCP_MD5SIG
if (key) {
*topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
@@ -84,15 +70,10 @@
fl.fl_ip_sport = t1->source;
security_skb_classify_flow(skb, &fl);
- /* Pass a socket to ip6_dst_lookup either it is for RST
- * Underlying function will use this to retrieve the network
- * namespace
- */
if (!ip6_dst_lookup(ctl_sk, &buff->dst, &fl)) {
if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
ip6_xmit(ctl_sk, buff, &fl, NULL, 0);
TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
- TCP_INC_STATS_BH(net, TCP_MIB_OUTRSTS);
return;
}
}
...which starts to be trivial to combine.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-10 05:42:01 +08:00
|
|
|
__be32 *topt;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-10-10 05:42:40 +08:00
|
|
|
if (ts)
|
|
|
|
tot_len += TCPOLEN_TSTAMP_ALIGNED;
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
if (key)
|
|
|
|
tot_len += TCPOLEN_MD5SIG_ALIGNED;
|
|
|
|
#endif
|
|
|
|
|
|
|
|
buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len,
|
2005-04-17 06:20:36 +08:00
|
|
|
GFP_ATOMIC);
|
2007-02-09 22:24:49 +08:00
|
|
|
if (buff == NULL)
|
|
|
|
return;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
skb_reserve(buff, MAX_HEADER + sizeof(struct ipv6hdr) + tot_len);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
t1 = (struct tcphdr *) skb_push(buff, tot_len);
|
2010-04-21 15:47:15 +08:00
|
|
|
skb_reset_transport_header(buff);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
/* Swap the send and the receive. */
|
|
|
|
memset(t1, 0, sizeof(*t1));
|
|
|
|
t1->dest = th->source;
|
|
|
|
t1->source = th->dest;
|
2006-11-15 11:07:45 +08:00
|
|
|
t1->doff = tot_len / 4;
|
2008-10-10 05:42:40 +08:00
|
|
|
t1->seq = htonl(seq);
|
|
|
|
t1->ack_seq = htonl(ack);
|
|
|
|
t1->ack = !rst || !th->ack;
|
|
|
|
t1->rst = rst;
|
|
|
|
t1->window = htons(win);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
tcpv6: convert opt[] -> topt in tcp_v6_send_reset
after this I get:
$ diff-funcs tcp_v6_send_reset tcp_ipv6.c tcp_ipv6.c tcp_v6_send_ack
--- tcp_ipv6.c:tcp_v6_send_reset()
+++ tcp_ipv6.c:tcp_v6_send_ack()
@@ -1,4 +1,5 @@
-static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
+static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win,
u32 ts,
+ struct tcp_md5sig_key *key)
{
struct tcphdr *th = tcp_hdr(skb), *t1;
struct sk_buff *buff;
@@ -7,31 +8,14 @@
struct sock *ctl_sk = net->ipv6.tcp_sk;
unsigned int tot_len = sizeof(struct tcphdr);
__be32 *topt;
-#ifdef CONFIG_TCP_MD5SIG
- struct tcp_md5sig_key *key;
-#endif
-
- if (th->rst)
- return;
-
- if (!ipv6_unicast_destination(skb))
- return;
+ if (ts)
+ tot_len += TCPOLEN_TSTAMP_ALIGNED;
#ifdef CONFIG_TCP_MD5SIG
- if (sk)
- key = tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->daddr);
- else
- key = NULL;
-
if (key)
tot_len += TCPOLEN_MD5SIG_ALIGNED;
#endif
- /*
- * We need to grab some memory, and put together an RST,
- * and then put it into the queue to be sent.
- */
-
buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len,
GFP_ATOMIC);
if (buff == NULL)
@@ -46,18 +30,20 @@
t1->dest = th->source;
t1->source = th->dest;
t1->doff = tot_len / 4;
- t1->rst = 1;
-
- if(th->ack) {
- t1->seq = th->ack_seq;
- } else {
- t1->ack = 1;
- t1->ack_seq = htonl(ntohl(th->seq) + th->syn + th->fin
- + skb->len - (th->doff<<2));
- }
+ t1->seq = htonl(seq);
+ t1->ack_seq = htonl(ack);
+ t1->ack = 1;
+ t1->window = htons(win);
topt = (__be32 *)(t1 + 1);
+ if (ts) {
+ *topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
+ (TCPOPT_TIMESTAMP << 8) |
TCPOLEN_TIMESTAMP);
+ *topt++ = htonl(tcp_time_stamp);
+ *topt++ = htonl(ts);
+ }
+
#ifdef CONFIG_TCP_MD5SIG
if (key) {
*topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
@@ -84,15 +70,10 @@
fl.fl_ip_sport = t1->source;
security_skb_classify_flow(skb, &fl);
- /* Pass a socket to ip6_dst_lookup either it is for RST
- * Underlying function will use this to retrieve the network
- * namespace
- */
if (!ip6_dst_lookup(ctl_sk, &buff->dst, &fl)) {
if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
ip6_xmit(ctl_sk, buff, &fl, NULL, 0);
TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
- TCP_INC_STATS_BH(net, TCP_MIB_OUTRSTS);
return;
}
}
...which starts to be trivial to combine.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-10 05:42:01 +08:00
|
|
|
topt = (__be32 *)(t1 + 1);
|
|
|
|
|
2008-10-10 05:42:40 +08:00
|
|
|
if (ts) {
|
|
|
|
*topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
|
|
|
|
(TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP);
|
|
|
|
*topt++ = htonl(tcp_time_stamp);
|
|
|
|
*topt++ = htonl(ts);
|
|
|
|
}
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
if (key) {
|
tcpv6: convert opt[] -> topt in tcp_v6_send_reset
after this I get:
$ diff-funcs tcp_v6_send_reset tcp_ipv6.c tcp_ipv6.c tcp_v6_send_ack
--- tcp_ipv6.c:tcp_v6_send_reset()
+++ tcp_ipv6.c:tcp_v6_send_ack()
@@ -1,4 +1,5 @@
-static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
+static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win,
u32 ts,
+ struct tcp_md5sig_key *key)
{
struct tcphdr *th = tcp_hdr(skb), *t1;
struct sk_buff *buff;
@@ -7,31 +8,14 @@
struct sock *ctl_sk = net->ipv6.tcp_sk;
unsigned int tot_len = sizeof(struct tcphdr);
__be32 *topt;
-#ifdef CONFIG_TCP_MD5SIG
- struct tcp_md5sig_key *key;
-#endif
-
- if (th->rst)
- return;
-
- if (!ipv6_unicast_destination(skb))
- return;
+ if (ts)
+ tot_len += TCPOLEN_TSTAMP_ALIGNED;
#ifdef CONFIG_TCP_MD5SIG
- if (sk)
- key = tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->daddr);
- else
- key = NULL;
-
if (key)
tot_len += TCPOLEN_MD5SIG_ALIGNED;
#endif
- /*
- * We need to grab some memory, and put together an RST,
- * and then put it into the queue to be sent.
- */
-
buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len,
GFP_ATOMIC);
if (buff == NULL)
@@ -46,18 +30,20 @@
t1->dest = th->source;
t1->source = th->dest;
t1->doff = tot_len / 4;
- t1->rst = 1;
-
- if(th->ack) {
- t1->seq = th->ack_seq;
- } else {
- t1->ack = 1;
- t1->ack_seq = htonl(ntohl(th->seq) + th->syn + th->fin
- + skb->len - (th->doff<<2));
- }
+ t1->seq = htonl(seq);
+ t1->ack_seq = htonl(ack);
+ t1->ack = 1;
+ t1->window = htons(win);
topt = (__be32 *)(t1 + 1);
+ if (ts) {
+ *topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
+ (TCPOPT_TIMESTAMP << 8) |
TCPOLEN_TIMESTAMP);
+ *topt++ = htonl(tcp_time_stamp);
+ *topt++ = htonl(ts);
+ }
+
#ifdef CONFIG_TCP_MD5SIG
if (key) {
*topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
@@ -84,15 +70,10 @@
fl.fl_ip_sport = t1->source;
security_skb_classify_flow(skb, &fl);
- /* Pass a socket to ip6_dst_lookup either it is for RST
- * Underlying function will use this to retrieve the network
- * namespace
- */
if (!ip6_dst_lookup(ctl_sk, &buff->dst, &fl)) {
if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
ip6_xmit(ctl_sk, buff, &fl, NULL, 0);
TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
- TCP_INC_STATS_BH(net, TCP_MIB_OUTRSTS);
return;
}
}
...which starts to be trivial to combine.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-10 05:42:01 +08:00
|
|
|
*topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
|
|
|
|
(TCPOPT_MD5SIG << 8) | TCPOLEN_MD5SIG);
|
|
|
|
tcp_v6_md5_hash_hdr((__u8 *)topt, key,
|
2008-10-10 05:37:47 +08:00
|
|
|
&ipv6_hdr(skb)->saddr,
|
|
|
|
&ipv6_hdr(skb)->daddr, t1);
|
2006-11-15 11:07:45 +08:00
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
memset(&fl6, 0, sizeof(fl6));
|
|
|
|
ipv6_addr_copy(&fl6.daddr, &ipv6_hdr(skb)->saddr);
|
|
|
|
ipv6_addr_copy(&fl6.saddr, &ipv6_hdr(skb)->daddr);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2010-04-22 05:59:20 +08:00
|
|
|
buff->ip_summed = CHECKSUM_PARTIAL;
|
|
|
|
buff->csum = 0;
|
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
__tcp_v6_send_check(buff, &fl6.saddr, &fl6.daddr);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2011-03-13 05:22:43 +08:00
|
|
|
fl6.flowi6_proto = IPPROTO_TCP;
|
|
|
|
fl6.flowi6_oif = inet6_iif(skb);
|
2011-03-13 05:36:19 +08:00
|
|
|
fl6.fl6_dport = t1->dest;
|
|
|
|
fl6.fl6_sport = t1->source;
|
2011-03-13 05:22:43 +08:00
|
|
|
security_skb_classify_flow(skb, flowi6_to_flowi(&fl6));
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-03-06 02:48:35 +08:00
|
|
|
/* Pass a socket to ip6_dst_lookup either it is for RST
|
|
|
|
* Underlying function will use this to retrieve the network
|
|
|
|
* namespace
|
|
|
|
*/
|
2011-03-13 05:22:43 +08:00
|
|
|
dst = ip6_dst_lookup_flow(ctl_sk, &fl6, NULL, false);
|
2011-03-02 05:19:07 +08:00
|
|
|
if (!IS_ERR(dst)) {
|
|
|
|
skb_dst_set(buff, dst);
|
2011-10-27 12:44:35 +08:00
|
|
|
ip6_xmit(ctl_sk, buff, &fl6, NULL, tclass);
|
2011-03-02 05:19:07 +08:00
|
|
|
TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
|
|
|
|
if (rst)
|
|
|
|
TCP_INC_STATS_BH(net, TCP_MIB_OUTRSTS);
|
|
|
|
return;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
kfree_skb(buff);
|
|
|
|
}
|
|
|
|
|
2008-10-10 05:42:40 +08:00
|
|
|
static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2011-10-21 17:22:42 +08:00
|
|
|
const struct tcphdr *th = tcp_hdr(skb);
|
2008-10-10 05:42:40 +08:00
|
|
|
u32 seq = 0, ack_seq = 0;
|
2008-10-10 12:11:56 +08:00
|
|
|
struct tcp_md5sig_key *key = NULL;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-10-10 05:42:40 +08:00
|
|
|
if (th->rst)
|
2005-04-17 06:20:36 +08:00
|
|
|
return;
|
|
|
|
|
2008-10-10 05:42:40 +08:00
|
|
|
if (!ipv6_unicast_destination(skb))
|
|
|
|
return;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
2008-10-10 05:42:40 +08:00
|
|
|
if (sk)
|
|
|
|
key = tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->daddr);
|
2006-11-15 11:07:45 +08:00
|
|
|
#endif
|
|
|
|
|
2008-10-10 05:42:40 +08:00
|
|
|
if (th->ack)
|
|
|
|
seq = ntohl(th->ack_seq);
|
|
|
|
else
|
|
|
|
ack_seq = ntohl(th->seq) + th->syn + th->fin + skb->len -
|
|
|
|
(th->doff << 2);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2011-10-27 12:44:35 +08:00
|
|
|
tcp_v6_send_response(skb, seq, ack_seq, 0, 0, key, 1, 0);
|
2008-10-10 05:42:40 +08:00
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-10-10 05:42:40 +08:00
|
|
|
static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts,
|
2011-10-27 12:44:35 +08:00
|
|
|
struct tcp_md5sig_key *key, u8 tclass)
|
2008-10-10 05:42:40 +08:00
|
|
|
{
|
2011-10-27 12:44:35 +08:00
|
|
|
tcp_v6_send_response(skb, seq, ack, win, ts, key, 0, tclass);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb)
|
|
|
|
{
|
2005-08-10 11:09:30 +08:00
|
|
|
struct inet_timewait_sock *tw = inet_twsk(sk);
|
2006-11-15 11:07:45 +08:00
|
|
|
struct tcp_timewait_sock *tcptw = tcp_twsk(sk);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-04-18 11:45:16 +08:00
|
|
|
tcp_v6_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
|
2005-08-10 11:09:30 +08:00
|
|
|
tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
|
2011-10-27 12:44:35 +08:00
|
|
|
tcptw->tw_ts_recent, tcp_twsk_md5_key(tcptw),
|
|
|
|
tw->tw_tclass);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2005-08-10 11:09:30 +08:00
|
|
|
inet_twsk_put(tw);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2008-08-07 14:50:04 +08:00
|
|
|
static void tcp_v6_reqsk_send_ack(struct sock *sk, struct sk_buff *skb,
|
|
|
|
struct request_sock *req)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2008-04-18 11:45:16 +08:00
|
|
|
tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent,
|
2011-10-27 12:44:35 +08:00
|
|
|
tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->daddr), 0);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb)
|
|
|
|
{
|
2005-06-19 13:47:21 +08:00
|
|
|
struct request_sock *req, **prev;
|
2007-04-11 12:04:22 +08:00
|
|
|
const struct tcphdr *th = tcp_hdr(skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
struct sock *nsk;
|
|
|
|
|
|
|
|
/* Find possible connection requests. */
|
2005-12-14 15:15:24 +08:00
|
|
|
req = inet6_csk_search_req(sk, &prev, th->source,
|
2007-04-26 08:54:47 +08:00
|
|
|
&ipv6_hdr(skb)->saddr,
|
|
|
|
&ipv6_hdr(skb)->daddr, inet6_iif(skb));
|
2005-04-17 06:20:36 +08:00
|
|
|
if (req)
|
|
|
|
return tcp_check_req(sk, skb, req, prev);
|
|
|
|
|
2008-03-26 01:26:21 +08:00
|
|
|
nsk = __inet6_lookup_established(sock_net(sk), &tcp_hashinfo,
|
2008-01-31 21:07:21 +08:00
|
|
|
&ipv6_hdr(skb)->saddr, th->source,
|
|
|
|
&ipv6_hdr(skb)->daddr, ntohs(th->dest), inet6_iif(skb));
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (nsk) {
|
|
|
|
if (nsk->sk_state != TCP_TIME_WAIT) {
|
|
|
|
bh_lock_sock(nsk);
|
|
|
|
return nsk;
|
|
|
|
}
|
2006-10-11 10:41:46 +08:00
|
|
|
inet_twsk_put(inet_twsk(nsk));
|
2005-04-17 06:20:36 +08:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2008-02-08 13:49:26 +08:00
|
|
|
#ifdef CONFIG_SYN_COOKIES
|
2010-06-03 08:43:44 +08:00
|
|
|
if (!th->syn)
|
2008-02-08 13:49:26 +08:00
|
|
|
sk = cookie_v6_check(sk, skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
#endif
|
|
|
|
return sk;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* FIXME: this is substantially similar to the ipv4 code.
|
|
|
|
* Can some kind of merge be done? -- erics
|
|
|
|
*/
|
|
|
|
static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
|
|
|
|
{
|
2009-12-03 02:25:27 +08:00
|
|
|
struct tcp_extend_values tmp_ext;
|
2009-12-03 02:07:39 +08:00
|
|
|
struct tcp_options_received tmp_opt;
|
2011-10-21 17:22:42 +08:00
|
|
|
const u8 *hash_location;
|
2009-12-03 02:07:39 +08:00
|
|
|
struct request_sock *req;
|
2005-12-14 15:15:40 +08:00
|
|
|
struct inet6_request_sock *treq;
|
2005-04-17 06:20:36 +08:00
|
|
|
struct ipv6_pinfo *np = inet6_sk(sk);
|
|
|
|
struct tcp_sock *tp = tcp_sk(sk);
|
2009-12-03 02:07:39 +08:00
|
|
|
__u32 isn = TCP_SKB_CB(skb)->when;
|
2010-12-03 04:14:29 +08:00
|
|
|
struct dst_entry *dst = NULL;
|
2008-02-08 13:49:26 +08:00
|
|
|
int want_cookie = 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (skb->protocol == htons(ETH_P_IP))
|
|
|
|
return tcp_v4_conn_request(sk, skb);
|
|
|
|
|
|
|
|
if (!ipv6_unicast_destination(skb))
|
2007-02-09 22:24:49 +08:00
|
|
|
goto drop;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2005-08-10 11:10:42 +08:00
|
|
|
if (inet_csk_reqsk_queue_is_full(sk) && !isn) {
|
2011-08-30 11:21:44 +08:00
|
|
|
want_cookie = tcp_syn_flood_action(sk, skb, "TCPv6");
|
|
|
|
if (!want_cookie)
|
|
|
|
goto drop;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2005-08-10 11:10:42 +08:00
|
|
|
if (sk_acceptq_is_full(sk) && inet_csk_reqsk_queue_young(sk) > 1)
|
2005-04-17 06:20:36 +08:00
|
|
|
goto drop;
|
|
|
|
|
2005-12-14 15:15:40 +08:00
|
|
|
req = inet6_reqsk_alloc(&tcp6_request_sock_ops);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (req == NULL)
|
|
|
|
goto drop;
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
tcp_rsk(req)->af_specific = &tcp_request_sock_ipv6_ops;
|
|
|
|
#endif
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
tcp_clear_options(&tmp_opt);
|
|
|
|
tmp_opt.mss_clamp = IPV6_MIN_MTU - sizeof(struct tcphdr) - sizeof(struct ipv6hdr);
|
|
|
|
tmp_opt.user_mss = tp->rx_opt.user_mss;
|
tcp: Revert per-route SACK/DSACK/TIMESTAMP changes.
It creates a regression, triggering badness for SYN_RECV
sockets, for example:
[19148.022102] Badness at net/ipv4/inet_connection_sock.c:293
[19148.022570] NIP: c02a0914 LR: c02a0904 CTR: 00000000
[19148.023035] REGS: eeecbd30 TRAP: 0700 Not tainted (2.6.32)
[19148.023496] MSR: 00029032 <EE,ME,CE,IR,DR> CR: 24002442 XER: 00000000
[19148.024012] TASK = eee9a820[1756] 'privoxy' THREAD: eeeca000
This is likely caused by the change in the 'estab' parameter
passed to tcp_parse_options() when invoked by the functions
in net/ipv4/tcp_minisocks.c
But even if that is fixed, the ->conn_request() changes made in
this patch series is fundamentally wrong. They try to use the
listening socket's 'dst' to probe the route settings. The
listening socket doesn't even have a route, and you can't
get the right route (the child request one) until much later
after we setup all of the state, and it must be done by hand.
This stuff really isn't ready, so the best thing to do is a
full revert. This reverts the following commits:
f55017a93f1a74d50244b1254b9a2bd7ac9bbf7d
022c3f7d82f0f1c68018696f2f027b87b9bb45c2
1aba721eba1d84a2defce45b950272cee1e6c72a
cda42ebd67ee5fdf09d7057b5a4584d36fe8a335
345cda2fd695534be5a4494f1b59da9daed33663
dc343475ed062e13fc260acccaab91d7d80fd5b2
05eaade2782fb0c90d3034fd7a7d5a16266182bb
6a2a2d6bf8581216e08be15fcb563cfd6c430e1e
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-16 12:56:42 +08:00
|
|
|
tcp_parse_options(skb, &tmp_opt, &hash_location, 0);
|
2009-12-03 02:25:27 +08:00
|
|
|
|
|
|
|
if (tmp_opt.cookie_plus > 0 &&
|
|
|
|
tmp_opt.saw_tstamp &&
|
|
|
|
!tp->rx_opt.cookie_out_never &&
|
|
|
|
(sysctl_tcp_cookie_size > 0 ||
|
|
|
|
(tp->cookie_values != NULL &&
|
|
|
|
tp->cookie_values->cookie_desired > 0))) {
|
|
|
|
u8 *c;
|
|
|
|
u32 *d;
|
|
|
|
u32 *mess = &tmp_ext.cookie_bakery[COOKIE_DIGEST_WORDS];
|
|
|
|
int l = tmp_opt.cookie_plus - TCPOLEN_COOKIE_BASE;
|
|
|
|
|
|
|
|
if (tcp_cookie_generator(&tmp_ext.cookie_bakery[0]) != 0)
|
|
|
|
goto drop_and_free;
|
|
|
|
|
|
|
|
/* Secret recipe starts with IP addresses */
|
2010-04-21 10:06:52 +08:00
|
|
|
d = (__force u32 *)&ipv6_hdr(skb)->daddr.s6_addr32[0];
|
2009-12-03 02:25:27 +08:00
|
|
|
*mess++ ^= *d++;
|
|
|
|
*mess++ ^= *d++;
|
|
|
|
*mess++ ^= *d++;
|
|
|
|
*mess++ ^= *d++;
|
2010-04-21 10:06:52 +08:00
|
|
|
d = (__force u32 *)&ipv6_hdr(skb)->saddr.s6_addr32[0];
|
2009-12-03 02:25:27 +08:00
|
|
|
*mess++ ^= *d++;
|
|
|
|
*mess++ ^= *d++;
|
|
|
|
*mess++ ^= *d++;
|
|
|
|
*mess++ ^= *d++;
|
|
|
|
|
|
|
|
/* plus variable length Initiator Cookie */
|
|
|
|
c = (u8 *)mess;
|
|
|
|
while (l-- > 0)
|
|
|
|
*c++ ^= *hash_location++;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2009-12-03 02:25:27 +08:00
|
|
|
want_cookie = 0; /* not our kind of cookie */
|
|
|
|
tmp_ext.cookie_out_never = 0; /* false */
|
|
|
|
tmp_ext.cookie_plus = tmp_opt.cookie_plus;
|
|
|
|
} else if (!tp->rx_opt.cookie_in_always) {
|
|
|
|
/* redundant indications, but ensure initialization. */
|
|
|
|
tmp_ext.cookie_out_never = 1; /* true */
|
|
|
|
tmp_ext.cookie_plus = 0;
|
|
|
|
} else {
|
|
|
|
goto drop_and_free;
|
|
|
|
}
|
|
|
|
tmp_ext.cookie_in_always = tp->rx_opt.cookie_in_always;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-04-10 18:12:40 +08:00
|
|
|
if (want_cookie && !tmp_opt.saw_tstamp)
|
2008-02-08 13:49:26 +08:00
|
|
|
tcp_clear_options(&tmp_opt);
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
tmp_opt.tstamp_ok = tmp_opt.saw_tstamp;
|
|
|
|
tcp_openreq_init(req, &tmp_opt, skb);
|
|
|
|
|
2005-12-14 15:15:40 +08:00
|
|
|
treq = inet6_rsk(req);
|
2007-04-26 08:54:47 +08:00
|
|
|
ipv6_addr_copy(&treq->rmt_addr, &ipv6_hdr(skb)->saddr);
|
|
|
|
ipv6_addr_copy(&treq->loc_addr, &ipv6_hdr(skb)->daddr);
|
2010-06-21 19:48:45 +08:00
|
|
|
if (!want_cookie || tmp_opt.tstamp_ok)
|
2008-02-08 13:49:26 +08:00
|
|
|
TCP_ECN_create_request(req, tcp_hdr(skb));
|
|
|
|
|
2011-11-24 06:29:23 +08:00
|
|
|
treq->iif = sk->sk_bound_dev_if;
|
|
|
|
|
|
|
|
/* So that link locals have meaning */
|
|
|
|
if (!sk->sk_bound_dev_if &&
|
|
|
|
ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL)
|
|
|
|
treq->iif = inet6_iif(skb);
|
|
|
|
|
2010-06-13 19:29:39 +08:00
|
|
|
if (!isn) {
|
2010-12-03 04:14:29 +08:00
|
|
|
struct inet_peer *peer = NULL;
|
|
|
|
|
2008-02-08 13:49:26 +08:00
|
|
|
if (ipv6_opt_accepted(sk, skb) ||
|
|
|
|
np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo ||
|
|
|
|
np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim) {
|
|
|
|
atomic_inc(&skb->users);
|
|
|
|
treq->pktopts = skb;
|
|
|
|
}
|
2010-12-03 04:14:29 +08:00
|
|
|
|
|
|
|
if (want_cookie) {
|
2010-06-13 19:29:39 +08:00
|
|
|
isn = cookie_v6_init_sequence(sk, skb, &req->mss);
|
|
|
|
req->cookie_ts = tmp_opt.tstamp_ok;
|
2010-12-03 04:14:29 +08:00
|
|
|
goto have_isn;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* VJ's idea. We save last timestamp seen
|
|
|
|
* from the destination in peer table, when entering
|
|
|
|
* state TIME-WAIT, and check against it before
|
|
|
|
* accepting new connection request.
|
|
|
|
*
|
|
|
|
* If "isn" is not zero, this request hit alive
|
|
|
|
* timewait bucket, so that all the necessary checks
|
|
|
|
* are made in the function processing timewait state.
|
|
|
|
*/
|
|
|
|
if (tmp_opt.saw_tstamp &&
|
|
|
|
tcp_death_row.sysctl_tw_recycle &&
|
|
|
|
(dst = inet6_csk_route_req(sk, req)) != NULL &&
|
|
|
|
(peer = rt6_get_peer((struct rt6_info *)dst)) != NULL &&
|
2011-02-10 06:30:26 +08:00
|
|
|
ipv6_addr_equal((struct in6_addr *)peer->daddr.addr.a6,
|
2010-12-03 04:14:29 +08:00
|
|
|
&treq->rmt_addr)) {
|
|
|
|
inet_peer_refcheck(peer);
|
|
|
|
if ((u32)get_seconds() - peer->tcp_ts_stamp < TCP_PAWS_MSL &&
|
|
|
|
(s32)(peer->tcp_ts - req->ts_recent) >
|
|
|
|
TCP_PAWS_WINDOW) {
|
|
|
|
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSPASSIVEREJECTED);
|
|
|
|
goto drop_and_release;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
/* Kill the following clause, if you dislike this way. */
|
|
|
|
else if (!sysctl_tcp_syncookies &&
|
|
|
|
(sysctl_max_syn_backlog - inet_csk_reqsk_queue_len(sk) <
|
|
|
|
(sysctl_max_syn_backlog >> 2)) &&
|
|
|
|
(!peer || !peer->tcp_ts_stamp) &&
|
|
|
|
(!dst || !dst_metric(dst, RTAX_RTT))) {
|
|
|
|
/* Without syncookies last quarter of
|
|
|
|
* backlog is filled with destinations,
|
|
|
|
* proven to be alive.
|
|
|
|
* It means that we continue to communicate
|
|
|
|
* to destinations, already remembered
|
|
|
|
* to the moment of synflood.
|
|
|
|
*/
|
|
|
|
LIMIT_NETDEBUG(KERN_DEBUG "TCP: drop open request from %pI6/%u\n",
|
|
|
|
&treq->rmt_addr, ntohs(tcp_hdr(skb)->source));
|
|
|
|
goto drop_and_release;
|
2010-06-13 19:29:39 +08:00
|
|
|
}
|
2010-12-03 04:14:29 +08:00
|
|
|
|
|
|
|
isn = tcp_v6_init_sequence(skb);
|
2008-02-08 13:49:26 +08:00
|
|
|
}
|
2010-12-03 04:14:29 +08:00
|
|
|
have_isn:
|
[NET] Generalise TCP's struct open_request minisock infrastructure
Kept this first changeset minimal, without changing existing names to
ease peer review.
Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn
has two new members:
->slab, that replaces tcp_openreq_cachep
->obj_size, to inform the size of the openreq descendant for
a specific protocol
The protocol specific fields in struct open_request were moved to a
class hierarchy, with the things that are common to all connection
oriented PF_INET protocols in struct inet_request_sock, the TCP ones
in tcp_request_sock, that is an inet_request_sock, that is an
open_request.
I.e. this uses the same approach used for the struct sock class
hierarchy, with sk_prot indicating if the protocol wants to use the
open_request infrastructure by filling in sk_prot->rsk_prot with an
or_calltable.
Results? Performance is improved and TCP v4 now uses only 64 bytes per
open request minisock, down from 96 without this patch :-)
Next changeset will rename some of the structs, fields and functions
mentioned above, struct or_calltable is way unclear, better name it
struct request_sock_ops, s/struct open_request/struct request_sock/g,
etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-19 13:46:52 +08:00
|
|
|
tcp_rsk(req)->snt_isn = isn;
|
2011-06-08 19:08:38 +08:00
|
|
|
tcp_rsk(req)->snt_synack = tcp_time_stamp;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-07-25 14:32:50 +08:00
|
|
|
security_inet_conn_request(sk, skb, req);
|
|
|
|
|
2009-12-03 02:25:27 +08:00
|
|
|
if (tcp_v6_send_synack(sk, req,
|
|
|
|
(struct request_values *)&tmp_ext) ||
|
|
|
|
want_cookie)
|
2009-12-03 02:07:39 +08:00
|
|
|
goto drop_and_free;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2009-12-03 02:07:39 +08:00
|
|
|
inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT);
|
|
|
|
return 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2010-12-03 04:14:29 +08:00
|
|
|
drop_and_release:
|
|
|
|
dst_release(dst);
|
2009-12-03 02:07:39 +08:00
|
|
|
drop_and_free:
|
|
|
|
reqsk_free(req);
|
2005-04-17 06:20:36 +08:00
|
|
|
drop:
|
|
|
|
return 0; /* don't send reset */
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
|
2005-06-19 13:47:21 +08:00
|
|
|
struct request_sock *req,
|
2005-04-17 06:20:36 +08:00
|
|
|
struct dst_entry *dst)
|
|
|
|
{
|
2008-09-13 07:17:43 +08:00
|
|
|
struct inet6_request_sock *treq;
|
2005-04-17 06:20:36 +08:00
|
|
|
struct ipv6_pinfo *newnp, *np = inet6_sk(sk);
|
|
|
|
struct tcp6_sock *newtcp6sk;
|
|
|
|
struct inet_sock *newinet;
|
|
|
|
struct tcp_sock *newtp;
|
|
|
|
struct sock *newsk;
|
|
|
|
struct ipv6_txoptions *opt;
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
struct tcp_md5sig_key *key;
|
|
|
|
#endif
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (skb->protocol == htons(ETH_P_IP)) {
|
|
|
|
/*
|
|
|
|
* v6 mapped
|
|
|
|
*/
|
|
|
|
|
|
|
|
newsk = tcp_v4_syn_recv_sock(sk, skb, req, dst);
|
|
|
|
|
2007-02-09 22:24:49 +08:00
|
|
|
if (newsk == NULL)
|
2005-04-17 06:20:36 +08:00
|
|
|
return NULL;
|
|
|
|
|
|
|
|
newtcp6sk = (struct tcp6_sock *)newsk;
|
|
|
|
inet_sk(newsk)->pinet6 = &newtcp6sk->inet6;
|
|
|
|
|
|
|
|
newinet = inet_sk(newsk);
|
|
|
|
newnp = inet6_sk(newsk);
|
|
|
|
newtp = tcp_sk(newsk);
|
|
|
|
|
|
|
|
memcpy(newnp, np, sizeof(struct ipv6_pinfo));
|
|
|
|
|
2009-10-15 14:30:45 +08:00
|
|
|
ipv6_addr_set_v4mapped(newinet->inet_daddr, &newnp->daddr);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2009-10-15 14:30:45 +08:00
|
|
|
ipv6_addr_set_v4mapped(newinet->inet_saddr, &newnp->saddr);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
ipv6_addr_copy(&newnp->rcv_saddr, &newnp->saddr);
|
|
|
|
|
2005-12-14 15:15:52 +08:00
|
|
|
inet_csk(newsk)->icsk_af_ops = &ipv6_mapped;
|
2005-04-17 06:20:36 +08:00
|
|
|
newsk->sk_backlog_rcv = tcp_v4_do_rcv;
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
newtp->af_specific = &tcp_sock_ipv6_mapped_specific;
|
|
|
|
#endif
|
|
|
|
|
2011-09-25 10:21:30 +08:00
|
|
|
newnp->ipv6_ac_list = NULL;
|
|
|
|
newnp->ipv6_fl_list = NULL;
|
2005-04-17 06:20:36 +08:00
|
|
|
newnp->pktoptions = NULL;
|
|
|
|
newnp->opt = NULL;
|
2005-08-12 20:19:38 +08:00
|
|
|
newnp->mcast_oif = inet6_iif(skb);
|
2007-04-26 08:54:47 +08:00
|
|
|
newnp->mcast_hops = ipv6_hdr(skb)->hop_limit;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2005-08-10 10:45:38 +08:00
|
|
|
/*
|
|
|
|
* No need to charge this sock to the relevant IPv6 refcnt debug socks count
|
|
|
|
* here, tcp_create_openreq_child now does this for us, see the comment in
|
|
|
|
* that function for the gory details. -acme
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
|
|
|
|
|
|
|
/* It is tricky place. Until this moment IPv4 tcp
|
2005-12-14 15:15:52 +08:00
|
|
|
worked with IPv6 icsk.icsk_af_ops.
|
2005-04-17 06:20:36 +08:00
|
|
|
Sync it now.
|
|
|
|
*/
|
2005-12-14 15:26:10 +08:00
|
|
|
tcp_sync_mss(newsk, inet_csk(newsk)->icsk_pmtu_cookie);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
return newsk;
|
|
|
|
}
|
|
|
|
|
2008-09-13 07:17:43 +08:00
|
|
|
treq = inet6_rsk(req);
|
2005-04-17 06:20:36 +08:00
|
|
|
opt = np->opt;
|
|
|
|
|
|
|
|
if (sk_acceptq_is_full(sk))
|
|
|
|
goto out_overflow;
|
|
|
|
|
2010-12-03 04:14:29 +08:00
|
|
|
if (!dst) {
|
|
|
|
dst = inet6_csk_route_req(sk, req);
|
|
|
|
if (!dst)
|
2005-04-17 06:20:36 +08:00
|
|
|
goto out;
|
2007-02-09 22:24:49 +08:00
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
newsk = tcp_create_openreq_child(sk, req, skb);
|
|
|
|
if (newsk == NULL)
|
2010-10-21 19:06:43 +08:00
|
|
|
goto out_nonewsk;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2005-08-10 10:45:38 +08:00
|
|
|
/*
|
|
|
|
* No need to charge this sock to the relevant IPv6 refcnt debug socks
|
|
|
|
* count here, tcp_create_openreq_child now does this for us, see the
|
|
|
|
* comment in that function for the gory details. -acme
|
|
|
|
*/
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-08-26 06:55:43 +08:00
|
|
|
newsk->sk_gso_type = SKB_GSO_TCPV6;
|
2006-08-30 08:15:09 +08:00
|
|
|
__ip6_dst_store(newsk, dst, NULL, NULL);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
newtcp6sk = (struct tcp6_sock *)newsk;
|
|
|
|
inet_sk(newsk)->pinet6 = &newtcp6sk->inet6;
|
|
|
|
|
|
|
|
newtp = tcp_sk(newsk);
|
|
|
|
newinet = inet_sk(newsk);
|
|
|
|
newnp = inet6_sk(newsk);
|
|
|
|
|
|
|
|
memcpy(newnp, np, sizeof(struct ipv6_pinfo));
|
|
|
|
|
[NET] Generalise TCP's struct open_request minisock infrastructure
Kept this first changeset minimal, without changing existing names to
ease peer review.
Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn
has two new members:
->slab, that replaces tcp_openreq_cachep
->obj_size, to inform the size of the openreq descendant for
a specific protocol
The protocol specific fields in struct open_request were moved to a
class hierarchy, with the things that are common to all connection
oriented PF_INET protocols in struct inet_request_sock, the TCP ones
in tcp_request_sock, that is an inet_request_sock, that is an
open_request.
I.e. this uses the same approach used for the struct sock class
hierarchy, with sk_prot indicating if the protocol wants to use the
open_request infrastructure by filling in sk_prot->rsk_prot with an
or_calltable.
Results? Performance is improved and TCP v4 now uses only 64 bytes per
open request minisock, down from 96 without this patch :-)
Next changeset will rename some of the structs, fields and functions
mentioned above, struct or_calltable is way unclear, better name it
struct request_sock_ops, s/struct open_request/struct request_sock/g,
etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-19 13:46:52 +08:00
|
|
|
ipv6_addr_copy(&newnp->daddr, &treq->rmt_addr);
|
|
|
|
ipv6_addr_copy(&newnp->saddr, &treq->loc_addr);
|
|
|
|
ipv6_addr_copy(&newnp->rcv_saddr, &treq->loc_addr);
|
|
|
|
newsk->sk_bound_dev_if = treq->iif;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2007-02-09 22:24:49 +08:00
|
|
|
/* Now IPv6 options...
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
First: no IPv4 options.
|
|
|
|
*/
|
2011-04-21 17:45:37 +08:00
|
|
|
newinet->inet_opt = NULL;
|
2011-09-25 10:21:30 +08:00
|
|
|
newnp->ipv6_ac_list = NULL;
|
2007-03-17 07:14:03 +08:00
|
|
|
newnp->ipv6_fl_list = NULL;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
/* Clone RX bits */
|
|
|
|
newnp->rxopt.all = np->rxopt.all;
|
|
|
|
|
|
|
|
/* Clone pktoptions received with SYN */
|
|
|
|
newnp->pktoptions = NULL;
|
[NET] Generalise TCP's struct open_request minisock infrastructure
Kept this first changeset minimal, without changing existing names to
ease peer review.
Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn
has two new members:
->slab, that replaces tcp_openreq_cachep
->obj_size, to inform the size of the openreq descendant for
a specific protocol
The protocol specific fields in struct open_request were moved to a
class hierarchy, with the things that are common to all connection
oriented PF_INET protocols in struct inet_request_sock, the TCP ones
in tcp_request_sock, that is an inet_request_sock, that is an
open_request.
I.e. this uses the same approach used for the struct sock class
hierarchy, with sk_prot indicating if the protocol wants to use the
open_request infrastructure by filling in sk_prot->rsk_prot with an
or_calltable.
Results? Performance is improved and TCP v4 now uses only 64 bytes per
open request minisock, down from 96 without this patch :-)
Next changeset will rename some of the structs, fields and functions
mentioned above, struct or_calltable is way unclear, better name it
struct request_sock_ops, s/struct open_request/struct request_sock/g,
etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-19 13:46:52 +08:00
|
|
|
if (treq->pktopts != NULL) {
|
|
|
|
newnp->pktoptions = skb_clone(treq->pktopts, GFP_ATOMIC);
|
|
|
|
kfree_skb(treq->pktopts);
|
|
|
|
treq->pktopts = NULL;
|
2005-04-17 06:20:36 +08:00
|
|
|
if (newnp->pktoptions)
|
|
|
|
skb_set_owner_r(newnp->pktoptions, newsk);
|
|
|
|
}
|
|
|
|
newnp->opt = NULL;
|
2005-08-12 20:19:38 +08:00
|
|
|
newnp->mcast_oif = inet6_iif(skb);
|
2007-04-26 08:54:47 +08:00
|
|
|
newnp->mcast_hops = ipv6_hdr(skb)->hop_limit;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
/* Clone native IPv6 options from listening socket (if any)
|
|
|
|
|
|
|
|
Yes, keeping reference count would be much more clever,
|
|
|
|
but we make one more one thing there: reattach optmem
|
|
|
|
to newsk.
|
|
|
|
*/
|
|
|
|
if (opt) {
|
|
|
|
newnp->opt = ipv6_dup_options(newsk, opt);
|
|
|
|
if (opt != np->opt)
|
|
|
|
sock_kfree_s(sk, opt, opt->tot_len);
|
|
|
|
}
|
|
|
|
|
2005-12-14 15:26:10 +08:00
|
|
|
inet_csk(newsk)->icsk_ext_hdr_len = 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
if (newnp->opt)
|
2005-12-14 15:26:10 +08:00
|
|
|
inet_csk(newsk)->icsk_ext_hdr_len = (newnp->opt->opt_nflen +
|
|
|
|
newnp->opt->opt_flen);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-03-21 09:53:41 +08:00
|
|
|
tcp_mtup_init(newsk);
|
2005-04-17 06:20:36 +08:00
|
|
|
tcp_sync_mss(newsk, dst_mtu(dst));
|
2010-12-14 04:52:14 +08:00
|
|
|
newtp->advmss = dst_metric_advmss(dst);
|
2005-04-17 06:20:36 +08:00
|
|
|
tcp_initialize_rcv_mss(newsk);
|
2011-06-08 19:08:38 +08:00
|
|
|
if (tcp_rsk(req)->snt_synack)
|
|
|
|
tcp_valid_rtt_meas(newsk,
|
|
|
|
tcp_time_stamp - tcp_rsk(req)->snt_synack);
|
|
|
|
newtp->total_retrans = req->retrans;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2009-10-15 14:30:45 +08:00
|
|
|
newinet->inet_daddr = newinet->inet_saddr = LOOPBACK4_IPV6;
|
|
|
|
newinet->inet_rcv_saddr = LOOPBACK4_IPV6;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
/* Copy over the MD5 key from the original socket */
|
|
|
|
if ((key = tcp_v6_md5_do_lookup(sk, &newnp->daddr)) != NULL) {
|
|
|
|
/* We're using one, so create a matching key
|
|
|
|
* on the newsk structure. If we fail to get
|
|
|
|
* memory, then we end up not copying the key
|
|
|
|
* across. Shucks.
|
|
|
|
*/
|
2006-11-17 22:14:37 +08:00
|
|
|
char *newkey = kmemdup(key->key, key->keylen, GFP_ATOMIC);
|
|
|
|
if (newkey != NULL)
|
2009-07-17 17:23:22 +08:00
|
|
|
tcp_v6_md5_do_add(newsk, &newnp->daddr,
|
2006-11-15 11:07:45 +08:00
|
|
|
newkey, key->keylen);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2010-10-21 19:06:43 +08:00
|
|
|
if (__inet_inherit_port(sk, newsk) < 0) {
|
|
|
|
sock_put(newsk);
|
|
|
|
goto out;
|
|
|
|
}
|
2009-12-04 11:46:54 +08:00
|
|
|
__inet6_hash(newsk, NULL);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
return newsk;
|
|
|
|
|
|
|
|
out_overflow:
|
2008-07-17 11:31:16 +08:00
|
|
|
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS);
|
2010-10-21 19:06:43 +08:00
|
|
|
out_nonewsk:
|
2005-04-17 06:20:36 +08:00
|
|
|
if (opt && opt != np->opt)
|
|
|
|
sock_kfree_s(sk, opt, opt->tot_len);
|
|
|
|
dst_release(dst);
|
2010-10-21 19:06:43 +08:00
|
|
|
out:
|
|
|
|
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS);
|
2005-04-17 06:20:36 +08:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2006-11-15 13:40:42 +08:00
|
|
|
static __sum16 tcp_v6_checksum_init(struct sk_buff *skb)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2006-08-30 07:44:56 +08:00
|
|
|
if (skb->ip_summed == CHECKSUM_COMPLETE) {
|
2009-01-09 02:41:23 +08:00
|
|
|
if (!tcp_v6_check(skb->len, &ipv6_hdr(skb)->saddr,
|
2007-04-26 08:54:47 +08:00
|
|
|
&ipv6_hdr(skb)->daddr, skb->csum)) {
|
2005-11-11 05:01:24 +08:00
|
|
|
skb->ip_summed = CHECKSUM_UNNECESSARY;
|
2005-04-17 06:20:36 +08:00
|
|
|
return 0;
|
2005-11-11 05:01:24 +08:00
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
2005-11-11 05:01:24 +08:00
|
|
|
|
2009-01-09 02:41:23 +08:00
|
|
|
skb->csum = ~csum_unfold(tcp_v6_check(skb->len,
|
2007-04-26 08:54:47 +08:00
|
|
|
&ipv6_hdr(skb)->saddr,
|
|
|
|
&ipv6_hdr(skb)->daddr, 0));
|
2005-11-11 05:01:24 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
if (skb->len <= 76) {
|
2005-11-11 05:01:24 +08:00
|
|
|
return __skb_checksum_complete(skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* The socket must have it's spinlock held when we get
|
|
|
|
* here.
|
|
|
|
*
|
|
|
|
* We have a potential double-lock case here, so even when
|
|
|
|
* doing backlog processing we use the BH locking scheme.
|
|
|
|
* This is because we cannot sleep with the original spinlock
|
|
|
|
* held.
|
|
|
|
*/
|
|
|
|
static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
struct ipv6_pinfo *np = inet6_sk(sk);
|
|
|
|
struct tcp_sock *tp;
|
|
|
|
struct sk_buff *opt_skb = NULL;
|
|
|
|
|
|
|
|
/* Imagine: socket is IPv6. IPv4 packet arrives,
|
|
|
|
goes to IPv4 receive handler and backlogged.
|
|
|
|
From backlog it always goes here. Kerboom...
|
|
|
|
Fortunately, tcp_rcv_established and rcv_established
|
|
|
|
handle them correctly, but it is not case with
|
|
|
|
tcp_v6_hnd_req and tcp_v6_send_reset(). --ANK
|
|
|
|
*/
|
|
|
|
|
|
|
|
if (skb->protocol == htons(ETH_P_IP))
|
|
|
|
return tcp_v4_do_rcv(sk, skb);
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
if (tcp_v6_inbound_md5_hash (sk, skb))
|
|
|
|
goto discard;
|
|
|
|
#endif
|
|
|
|
|
2006-09-01 06:28:39 +08:00
|
|
|
if (sk_filter(sk, skb))
|
2005-04-17 06:20:36 +08:00
|
|
|
goto discard;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* socket locking is here for SMP purposes as backlog rcv
|
|
|
|
* is currently called with bh processing disabled.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* Do Stevens' IPV6_PKTOPTIONS.
|
|
|
|
|
|
|
|
Yes, guys, it is the only place in our code, where we
|
|
|
|
may make it not affecting IPv4.
|
|
|
|
The rest of code is protocol independent,
|
|
|
|
and I do not like idea to uglify IPv4.
|
|
|
|
|
|
|
|
Actually, all the idea behind IPV6_PKTOPTIONS
|
|
|
|
looks not very well thought. For now we latch
|
|
|
|
options, received in the last packet, enqueued
|
|
|
|
by tcp. Feel free to propose better solution.
|
2007-02-09 22:24:49 +08:00
|
|
|
--ANK (980728)
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
|
|
|
if (np->rxopt.all)
|
|
|
|
opt_skb = skb_clone(skb, GFP_ATOMIC);
|
|
|
|
|
|
|
|
if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */
|
2011-08-15 03:45:55 +08:00
|
|
|
sock_rps_save_rxhash(sk, skb);
|
2007-04-11 12:04:22 +08:00
|
|
|
if (tcp_rcv_established(sk, skb, tcp_hdr(skb), skb->len))
|
2005-04-17 06:20:36 +08:00
|
|
|
goto reset;
|
|
|
|
if (opt_skb)
|
|
|
|
goto ipv6_pktoptions;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-03-19 08:43:48 +08:00
|
|
|
if (skb->len < tcp_hdrlen(skb) || tcp_checksum_complete(skb))
|
2005-04-17 06:20:36 +08:00
|
|
|
goto csum_err;
|
|
|
|
|
2007-02-09 22:24:49 +08:00
|
|
|
if (sk->sk_state == TCP_LISTEN) {
|
2005-04-17 06:20:36 +08:00
|
|
|
struct sock *nsk = tcp_v6_hnd_req(sk, skb);
|
|
|
|
if (!nsk)
|
|
|
|
goto discard;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Queue it on the new socket if the new socket is active,
|
|
|
|
* otherwise we just shortcircuit this and continue with
|
|
|
|
* the new socket..
|
|
|
|
*/
|
2007-02-09 22:24:49 +08:00
|
|
|
if(nsk != sk) {
|
2011-08-15 03:45:55 +08:00
|
|
|
sock_rps_save_rxhash(nsk, skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (tcp_child_process(sk, nsk, skb))
|
|
|
|
goto reset;
|
|
|
|
if (opt_skb)
|
|
|
|
__kfree_skb(opt_skb);
|
|
|
|
return 0;
|
|
|
|
}
|
2011-04-07 04:07:09 +08:00
|
|
|
} else
|
2011-08-15 03:45:55 +08:00
|
|
|
sock_rps_save_rxhash(sk, skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2007-04-11 12:04:22 +08:00
|
|
|
if (tcp_rcv_state_process(sk, skb, tcp_hdr(skb), skb->len))
|
2005-04-17 06:20:36 +08:00
|
|
|
goto reset;
|
|
|
|
if (opt_skb)
|
|
|
|
goto ipv6_pktoptions;
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
reset:
|
2006-11-15 11:07:45 +08:00
|
|
|
tcp_v6_send_reset(sk, skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
discard:
|
|
|
|
if (opt_skb)
|
|
|
|
__kfree_skb(opt_skb);
|
|
|
|
kfree_skb(skb);
|
|
|
|
return 0;
|
|
|
|
csum_err:
|
2008-07-17 11:22:25 +08:00
|
|
|
TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
|
2005-04-17 06:20:36 +08:00
|
|
|
goto discard;
|
|
|
|
|
|
|
|
|
|
|
|
ipv6_pktoptions:
|
|
|
|
/* Do you ask, what is it?
|
|
|
|
|
|
|
|
1. skb was enqueued by tcp.
|
|
|
|
2. skb is added to tail of read queue, rather than out of order.
|
|
|
|
3. socket is not in passive state.
|
|
|
|
4. Finally, it really contains options, which user wants to receive.
|
|
|
|
*/
|
|
|
|
tp = tcp_sk(sk);
|
|
|
|
if (TCP_SKB_CB(opt_skb)->end_seq == tp->rcv_nxt &&
|
|
|
|
!((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN))) {
|
[IPV6]: Support several new sockopt / ancillary data in Advanced API (RFC3542).
Support several new socket options / ancillary data:
IPV6_RECVPKTINFO, IPV6_PKTINFO,
IPV6_RECVHOPOPTS, IPV6_HOPOPTS,
IPV6_RECVDSTOPTS, IPV6_DSTOPTS, IPV6_RTHDRDSTOPTS,
IPV6_RECVRTHDR, IPV6_RTHDR,
IPV6_RECVHOPOPTS, IPV6_HOPOPTS
Old semantics are preserved as IPV6_2292xxxx so that
we can maintain backward compatibility.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
2005-09-08 08:59:17 +08:00
|
|
|
if (np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo)
|
2005-08-12 20:19:38 +08:00
|
|
|
np->mcast_oif = inet6_iif(opt_skb);
|
[IPV6]: Support several new sockopt / ancillary data in Advanced API (RFC3542).
Support several new socket options / ancillary data:
IPV6_RECVPKTINFO, IPV6_PKTINFO,
IPV6_RECVHOPOPTS, IPV6_HOPOPTS,
IPV6_RECVDSTOPTS, IPV6_DSTOPTS, IPV6_RTHDRDSTOPTS,
IPV6_RECVRTHDR, IPV6_RTHDR,
IPV6_RECVHOPOPTS, IPV6_HOPOPTS
Old semantics are preserved as IPV6_2292xxxx so that
we can maintain backward compatibility.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
2005-09-08 08:59:17 +08:00
|
|
|
if (np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim)
|
2007-04-26 08:54:47 +08:00
|
|
|
np->mcast_hops = ipv6_hdr(opt_skb)->hop_limit;
|
2005-04-17 06:20:36 +08:00
|
|
|
if (ipv6_opt_accepted(sk, opt_skb)) {
|
|
|
|
skb_set_owner_r(opt_skb, sk);
|
|
|
|
opt_skb = xchg(&np->pktoptions, opt_skb);
|
|
|
|
} else {
|
|
|
|
__kfree_skb(opt_skb);
|
|
|
|
opt_skb = xchg(&np->pktoptions, NULL);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-02-24 05:45:33 +08:00
|
|
|
kfree_skb(opt_skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-10-16 03:50:28 +08:00
|
|
|
static int tcp_v6_rcv(struct sk_buff *skb)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2011-10-21 17:22:42 +08:00
|
|
|
const struct tcphdr *th;
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct ipv6hdr *hdr;
|
2005-04-17 06:20:36 +08:00
|
|
|
struct sock *sk;
|
|
|
|
int ret;
|
2008-07-17 11:20:58 +08:00
|
|
|
struct net *net = dev_net(skb->dev);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (skb->pkt_type != PACKET_HOST)
|
|
|
|
goto discard_it;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Count it even if it's bad.
|
|
|
|
*/
|
2008-07-17 11:22:25 +08:00
|
|
|
TCP_INC_STATS_BH(net, TCP_MIB_INSEGS);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (!pskb_may_pull(skb, sizeof(struct tcphdr)))
|
|
|
|
goto discard_it;
|
|
|
|
|
2007-04-11 12:04:22 +08:00
|
|
|
th = tcp_hdr(skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (th->doff < sizeof(struct tcphdr)/4)
|
|
|
|
goto bad_packet;
|
|
|
|
if (!pskb_may_pull(skb, th->doff*4))
|
|
|
|
goto discard_it;
|
|
|
|
|
2007-04-10 02:59:39 +08:00
|
|
|
if (!skb_csum_unnecessary(skb) && tcp_v6_checksum_init(skb))
|
2005-04-17 06:20:36 +08:00
|
|
|
goto bad_packet;
|
|
|
|
|
2007-04-11 12:04:22 +08:00
|
|
|
th = tcp_hdr(skb);
|
IPv6: Generic TTL Security Mechanism (final version)
This patch adds IPv6 support for RFC5082 Generalized TTL Security Mechanism.
Not to users of mapped address; the IPV6 and IPV4 socket options are seperate.
The server does have to deal with both IPv4 and IPv6 socket options
and the client has to handle the different for each family.
On client:
int ttl = 255;
getaddrinfo(argv[1], argv[2], &hint, &result);
for (rp = result; rp != NULL; rp = rp->ai_next) {
s = socket(rp->ai_family, rp->ai_socktype, rp->ai_protocol);
if (s < 0) continue;
if (rp->ai_family == AF_INET) {
setsockopt(s, IPPROTO_IP, IP_TTL, &ttl, sizeof(ttl));
} else if (rp->ai_family == AF_INET6) {
setsockopt(s, IPPROTO_IPV6, IPV6_UNICAST_HOPS,
&ttl, sizeof(ttl)))
}
if (connect(s, rp->ai_addr, rp->ai_addrlen) == 0) {
...
On server:
int minttl = 255 - maxhops;
getaddrinfo(NULL, port, &hints, &result);
for (rp = result; rp != NULL; rp = rp->ai_next) {
s = socket(rp->ai_family, rp->ai_socktype, rp->ai_protocol);
if (s < 0) continue;
if (rp->ai_family == AF_INET6)
setsockopt(s, IPPROTO_IPV6, IPV6_MINHOPCOUNT,
&minttl, sizeof(minttl));
setsockopt(s, IPPROTO_IP, IP_MINTTL, &minttl, sizeof(minttl));
if (bind(s, rp->ai_addr, rp->ai_addrlen) == 0)
break
...
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-23 06:24:53 +08:00
|
|
|
hdr = ipv6_hdr(skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
TCP_SKB_CB(skb)->seq = ntohl(th->seq);
|
|
|
|
TCP_SKB_CB(skb)->end_seq = (TCP_SKB_CB(skb)->seq + th->syn + th->fin +
|
|
|
|
skb->len - th->doff*4);
|
|
|
|
TCP_SKB_CB(skb)->ack_seq = ntohl(th->ack_seq);
|
|
|
|
TCP_SKB_CB(skb)->when = 0;
|
2011-09-27 14:20:08 +08:00
|
|
|
TCP_SKB_CB(skb)->ip_dsfield = ipv6_get_dsfield(hdr);
|
2005-04-17 06:20:36 +08:00
|
|
|
TCP_SKB_CB(skb)->sacked = 0;
|
|
|
|
|
2008-10-08 02:41:57 +08:00
|
|
|
sk = __inet6_lookup_skb(&tcp_hashinfo, skb, th->source, th->dest);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (!sk)
|
|
|
|
goto no_tcp_socket;
|
|
|
|
|
|
|
|
process:
|
|
|
|
if (sk->sk_state == TCP_TIME_WAIT)
|
|
|
|
goto do_time_wait;
|
|
|
|
|
IPv6: Generic TTL Security Mechanism (final version)
This patch adds IPv6 support for RFC5082 Generalized TTL Security Mechanism.
Not to users of mapped address; the IPV6 and IPV4 socket options are seperate.
The server does have to deal with both IPv4 and IPv6 socket options
and the client has to handle the different for each family.
On client:
int ttl = 255;
getaddrinfo(argv[1], argv[2], &hint, &result);
for (rp = result; rp != NULL; rp = rp->ai_next) {
s = socket(rp->ai_family, rp->ai_socktype, rp->ai_protocol);
if (s < 0) continue;
if (rp->ai_family == AF_INET) {
setsockopt(s, IPPROTO_IP, IP_TTL, &ttl, sizeof(ttl));
} else if (rp->ai_family == AF_INET6) {
setsockopt(s, IPPROTO_IPV6, IPV6_UNICAST_HOPS,
&ttl, sizeof(ttl)))
}
if (connect(s, rp->ai_addr, rp->ai_addrlen) == 0) {
...
On server:
int minttl = 255 - maxhops;
getaddrinfo(NULL, port, &hints, &result);
for (rp = result; rp != NULL; rp = rp->ai_next) {
s = socket(rp->ai_family, rp->ai_socktype, rp->ai_protocol);
if (s < 0) continue;
if (rp->ai_family == AF_INET6)
setsockopt(s, IPPROTO_IPV6, IPV6_MINHOPCOUNT,
&minttl, sizeof(minttl));
setsockopt(s, IPPROTO_IP, IP_MINTTL, &minttl, sizeof(minttl));
if (bind(s, rp->ai_addr, rp->ai_addrlen) == 0)
break
...
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-23 06:24:53 +08:00
|
|
|
if (hdr->hop_limit < inet6_sk(sk)->min_hopcount) {
|
|
|
|
NET_INC_STATS_BH(net, LINUX_MIB_TCPMINTTLDROP);
|
|
|
|
goto discard_and_relse;
|
|
|
|
}
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb))
|
|
|
|
goto discard_and_relse;
|
|
|
|
|
2006-09-01 06:28:39 +08:00
|
|
|
if (sk_filter(sk, skb))
|
2005-04-17 06:20:36 +08:00
|
|
|
goto discard_and_relse;
|
|
|
|
|
|
|
|
skb->dev = NULL;
|
|
|
|
|
2006-09-26 13:28:47 +08:00
|
|
|
bh_lock_sock_nested(sk);
|
2005-04-17 06:20:36 +08:00
|
|
|
ret = 0;
|
|
|
|
if (!sock_owned_by_user(sk)) {
|
2006-05-24 09:05:53 +08:00
|
|
|
#ifdef CONFIG_NET_DMA
|
2007-02-09 22:24:49 +08:00
|
|
|
struct tcp_sock *tp = tcp_sk(sk);
|
2007-10-26 19:20:13 +08:00
|
|
|
if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
|
2009-01-07 02:38:15 +08:00
|
|
|
tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY);
|
2007-02-09 22:24:49 +08:00
|
|
|
if (tp->ucopy.dma_chan)
|
|
|
|
ret = tcp_v6_do_rcv(sk, skb);
|
|
|
|
else
|
2006-05-24 09:05:53 +08:00
|
|
|
#endif
|
|
|
|
{
|
|
|
|
if (!tcp_prequeue(sk, skb))
|
|
|
|
ret = tcp_v6_do_rcv(sk, skb);
|
|
|
|
}
|
2010-03-08 07:21:57 +08:00
|
|
|
} else if (unlikely(sk_add_backlog(sk, skb))) {
|
2010-03-05 02:01:41 +08:00
|
|
|
bh_unlock_sock(sk);
|
2010-03-08 07:21:57 +08:00
|
|
|
NET_INC_STATS_BH(net, LINUX_MIB_TCPBACKLOGDROP);
|
2010-03-05 02:01:41 +08:00
|
|
|
goto discard_and_relse;
|
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
bh_unlock_sock(sk);
|
|
|
|
|
|
|
|
sock_put(sk);
|
|
|
|
return ret ? -1 : 0;
|
|
|
|
|
|
|
|
no_tcp_socket:
|
|
|
|
if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb))
|
|
|
|
goto discard_it;
|
|
|
|
|
|
|
|
if (skb->len < (th->doff<<2) || tcp_checksum_complete(skb)) {
|
|
|
|
bad_packet:
|
2008-07-17 11:22:25 +08:00
|
|
|
TCP_INC_STATS_BH(net, TCP_MIB_INERRS);
|
2005-04-17 06:20:36 +08:00
|
|
|
} else {
|
2006-11-15 11:07:45 +08:00
|
|
|
tcp_v6_send_reset(NULL, skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
discard_it:
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Discard frame
|
|
|
|
*/
|
|
|
|
|
|
|
|
kfree_skb(skb);
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
discard_and_relse:
|
|
|
|
sock_put(sk);
|
|
|
|
goto discard_it;
|
|
|
|
|
|
|
|
do_time_wait:
|
|
|
|
if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) {
|
2006-10-11 10:41:46 +08:00
|
|
|
inet_twsk_put(inet_twsk(sk));
|
2005-04-17 06:20:36 +08:00
|
|
|
goto discard_it;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (skb->len < (th->doff<<2) || tcp_checksum_complete(skb)) {
|
2008-07-17 11:22:25 +08:00
|
|
|
TCP_INC_STATS_BH(net, TCP_MIB_INERRS);
|
2006-10-11 10:41:46 +08:00
|
|
|
inet_twsk_put(inet_twsk(sk));
|
2005-04-17 06:20:36 +08:00
|
|
|
goto discard_it;
|
|
|
|
}
|
|
|
|
|
2006-10-11 10:41:46 +08:00
|
|
|
switch (tcp_timewait_state_process(inet_twsk(sk), skb, th)) {
|
2005-04-17 06:20:36 +08:00
|
|
|
case TCP_TW_SYN:
|
|
|
|
{
|
|
|
|
struct sock *sk2;
|
|
|
|
|
2008-03-25 20:47:49 +08:00
|
|
|
sk2 = inet6_lookup_listener(dev_net(skb->dev), &tcp_hashinfo,
|
2007-04-26 08:54:47 +08:00
|
|
|
&ipv6_hdr(skb)->daddr,
|
2005-08-12 20:19:38 +08:00
|
|
|
ntohs(th->dest), inet6_iif(skb));
|
2005-04-17 06:20:36 +08:00
|
|
|
if (sk2 != NULL) {
|
2005-08-10 11:44:40 +08:00
|
|
|
struct inet_timewait_sock *tw = inet_twsk(sk);
|
|
|
|
inet_twsk_deschedule(tw, &tcp_death_row);
|
|
|
|
inet_twsk_put(tw);
|
2005-04-17 06:20:36 +08:00
|
|
|
sk = sk2;
|
|
|
|
goto process;
|
|
|
|
}
|
|
|
|
/* Fall through to ACK */
|
|
|
|
}
|
|
|
|
case TCP_TW_ACK:
|
|
|
|
tcp_v6_timewait_ack(sk, skb);
|
|
|
|
break;
|
|
|
|
case TCP_TW_RST:
|
|
|
|
goto no_tcp_socket;
|
|
|
|
case TCP_TW_SUCCESS:;
|
|
|
|
}
|
|
|
|
goto discard_it;
|
|
|
|
}
|
|
|
|
|
2010-12-02 10:09:13 +08:00
|
|
|
static struct inet_peer *tcp_v6_get_peer(struct sock *sk, bool *release_it)
|
|
|
|
{
|
2010-12-03 03:52:07 +08:00
|
|
|
struct rt6_info *rt = (struct rt6_info *) __sk_dst_get(sk);
|
|
|
|
struct ipv6_pinfo *np = inet6_sk(sk);
|
|
|
|
struct inet_peer *peer;
|
|
|
|
|
|
|
|
if (!rt ||
|
|
|
|
!ipv6_addr_equal(&np->daddr, &rt->rt6i_dst.addr)) {
|
|
|
|
peer = inet_getpeer_v6(&np->daddr, 1);
|
|
|
|
*release_it = true;
|
|
|
|
} else {
|
|
|
|
if (!rt->rt6i_peer)
|
|
|
|
rt6_bind_peer(rt, 1);
|
|
|
|
peer = rt->rt6i_peer;
|
2010-12-11 05:16:09 +08:00
|
|
|
*release_it = false;
|
2010-12-03 03:52:07 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
return peer;
|
2010-12-02 10:09:13 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static void *tcp_v6_tw_get_peer(struct sock *sk)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2011-10-21 17:22:42 +08:00
|
|
|
const struct inet6_timewait_sock *tw6 = inet6_twsk(sk);
|
|
|
|
const struct inet_timewait_sock *tw = inet_twsk(sk);
|
2010-12-02 10:09:13 +08:00
|
|
|
|
|
|
|
if (tw->tw_family == AF_INET)
|
|
|
|
return tcp_v4_tw_get_peer(sk);
|
|
|
|
|
2010-12-03 03:52:07 +08:00
|
|
|
return inet_getpeer_v6(&tw6->tw_v6_daddr, 1);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2010-12-02 10:09:13 +08:00
|
|
|
static struct timewait_sock_ops tcp6_timewait_sock_ops = {
|
|
|
|
.twsk_obj_size = sizeof(struct tcp6_timewait_sock),
|
|
|
|
.twsk_unique = tcp_twsk_unique,
|
|
|
|
.twsk_destructor= tcp_twsk_destructor,
|
|
|
|
.twsk_getpeer = tcp_v6_tw_get_peer,
|
|
|
|
};
|
|
|
|
|
2009-09-02 03:25:04 +08:00
|
|
|
static const struct inet_connection_sock_af_ops ipv6_specific = {
|
2006-03-21 14:48:35 +08:00
|
|
|
.queue_xmit = inet6_csk_xmit,
|
|
|
|
.send_check = tcp_v6_send_check,
|
|
|
|
.rebuild_header = inet6_sk_rebuild_header,
|
|
|
|
.conn_request = tcp_v6_conn_request,
|
|
|
|
.syn_recv_sock = tcp_v6_syn_recv_sock,
|
2010-11-30 05:37:14 +08:00
|
|
|
.get_peer = tcp_v6_get_peer,
|
2006-03-21 14:48:35 +08:00
|
|
|
.net_header_len = sizeof(struct ipv6hdr),
|
|
|
|
.setsockopt = ipv6_setsockopt,
|
|
|
|
.getsockopt = ipv6_getsockopt,
|
|
|
|
.addr2sockaddr = inet6_csk_addr2sockaddr,
|
|
|
|
.sockaddr_len = sizeof(struct sockaddr_in6),
|
[SOCK] proto: Add hashinfo member to struct proto
This way we can remove TCP and DCCP specific versions of
sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port
sk->sk_prot->hash: inet_hash is directly used, only v6 need
a specific version to deal with mapped sockets
sk->sk_prot->unhash: both v4 and v6 use inet_hash directly
struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so
that inet_csk_get_port can find the per family routine.
Now only the lookup routines receive as a parameter a struct inet_hashtable.
With this we further reuse code, reducing the difference among INET transport
protocols.
Eventually work has to be done on UDP and SCTP to make them share this
infrastructure and get as a bonus inet_diag interfaces so that iproute can be
used with these protocols.
net-2.6/net/ipv4/inet_hashtables.c:
struct proto | +8
struct inet_connection_sock_af_ops | +8
2 structs changed
__inet_hash_nolisten | +18
__inet_hash | -210
inet_put_port | +8
inet_bind_bucket_create | +1
__inet_hash_connect | -8
5 functions changed, 27 bytes added, 218 bytes removed, diff: -191
net-2.6/net/core/sock.c:
proto_seq_show | +3
1 function changed, 3 bytes added, diff: +3
net-2.6/net/ipv4/inet_connection_sock.c:
inet_csk_get_port | +15
1 function changed, 15 bytes added, diff: +15
net-2.6/net/ipv4/tcp.c:
tcp_set_state | -7
1 function changed, 7 bytes removed, diff: -7
net-2.6/net/ipv4/tcp_ipv4.c:
tcp_v4_get_port | -31
tcp_v4_hash | -48
tcp_v4_destroy_sock | -7
tcp_v4_syn_recv_sock | -2
tcp_unhash | -179
5 functions changed, 267 bytes removed, diff: -267
net-2.6/net/ipv6/inet6_hashtables.c:
__inet6_hash | +8
1 function changed, 8 bytes added, diff: +8
net-2.6/net/ipv4/inet_hashtables.c:
inet_unhash | +190
inet_hash | +242
2 functions changed, 432 bytes added, diff: +432
vmlinux:
16 functions changed, 485 bytes added, 492 bytes removed, diff: -7
/home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c:
tcp_v6_get_port | -31
tcp_v6_hash | -7
tcp_v6_syn_recv_sock | -9
3 functions changed, 47 bytes removed, diff: -47
/home/acme/git/net-2.6/net/dccp/proto.c:
dccp_destroy_sock | -7
dccp_unhash | -179
dccp_hash | -49
dccp_set_state | -7
dccp_done | +1
5 functions changed, 1 bytes added, 242 bytes removed, diff: -241
/home/acme/git/net-2.6/net/dccp/ipv4.c:
dccp_v4_get_port | -31
dccp_v4_request_recv_sock | -2
2 functions changed, 33 bytes removed, diff: -33
/home/acme/git/net-2.6/net/dccp/ipv6.c:
dccp_v6_get_port | -31
dccp_v6_hash | -7
dccp_v6_request_recv_sock | +5
3 functions changed, 5 bytes added, 38 bytes removed, diff: -33
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 20:06:04 +08:00
|
|
|
.bind_conflict = inet6_csk_bind_conflict,
|
2006-03-21 14:45:21 +08:00
|
|
|
#ifdef CONFIG_COMPAT
|
2006-03-21 14:48:35 +08:00
|
|
|
.compat_setsockopt = compat_ipv6_setsockopt,
|
|
|
|
.compat_getsockopt = compat_ipv6_getsockopt,
|
2006-03-21 14:45:21 +08:00
|
|
|
#endif
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
2009-09-02 03:25:03 +08:00
|
|
|
static const struct tcp_sock_af_ops tcp_sock_ipv6_specific = {
|
2006-11-15 11:07:45 +08:00
|
|
|
.md5_lookup = tcp_v6_md5_lookup,
|
2008-07-19 15:01:42 +08:00
|
|
|
.calc_md5_hash = tcp_v6_md5_hash_skb,
|
2006-11-15 11:07:45 +08:00
|
|
|
.md5_add = tcp_v6_md5_add_func,
|
|
|
|
.md5_parse = tcp_v6_parse_md5_keys,
|
|
|
|
};
|
2006-11-15 11:53:22 +08:00
|
|
|
#endif
|
2006-11-15 11:07:45 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/*
|
|
|
|
* TCP over IPv4 via INET6 API
|
|
|
|
*/
|
|
|
|
|
2009-09-02 03:25:04 +08:00
|
|
|
static const struct inet_connection_sock_af_ops ipv6_mapped = {
|
2006-03-21 14:48:35 +08:00
|
|
|
.queue_xmit = ip_queue_xmit,
|
|
|
|
.send_check = tcp_v4_send_check,
|
|
|
|
.rebuild_header = inet_sk_rebuild_header,
|
|
|
|
.conn_request = tcp_v6_conn_request,
|
|
|
|
.syn_recv_sock = tcp_v6_syn_recv_sock,
|
2010-11-30 05:37:14 +08:00
|
|
|
.get_peer = tcp_v4_get_peer,
|
2006-03-21 14:48:35 +08:00
|
|
|
.net_header_len = sizeof(struct iphdr),
|
|
|
|
.setsockopt = ipv6_setsockopt,
|
|
|
|
.getsockopt = ipv6_getsockopt,
|
|
|
|
.addr2sockaddr = inet6_csk_addr2sockaddr,
|
|
|
|
.sockaddr_len = sizeof(struct sockaddr_in6),
|
[SOCK] proto: Add hashinfo member to struct proto
This way we can remove TCP and DCCP specific versions of
sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port
sk->sk_prot->hash: inet_hash is directly used, only v6 need
a specific version to deal with mapped sockets
sk->sk_prot->unhash: both v4 and v6 use inet_hash directly
struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so
that inet_csk_get_port can find the per family routine.
Now only the lookup routines receive as a parameter a struct inet_hashtable.
With this we further reuse code, reducing the difference among INET transport
protocols.
Eventually work has to be done on UDP and SCTP to make them share this
infrastructure and get as a bonus inet_diag interfaces so that iproute can be
used with these protocols.
net-2.6/net/ipv4/inet_hashtables.c:
struct proto | +8
struct inet_connection_sock_af_ops | +8
2 structs changed
__inet_hash_nolisten | +18
__inet_hash | -210
inet_put_port | +8
inet_bind_bucket_create | +1
__inet_hash_connect | -8
5 functions changed, 27 bytes added, 218 bytes removed, diff: -191
net-2.6/net/core/sock.c:
proto_seq_show | +3
1 function changed, 3 bytes added, diff: +3
net-2.6/net/ipv4/inet_connection_sock.c:
inet_csk_get_port | +15
1 function changed, 15 bytes added, diff: +15
net-2.6/net/ipv4/tcp.c:
tcp_set_state | -7
1 function changed, 7 bytes removed, diff: -7
net-2.6/net/ipv4/tcp_ipv4.c:
tcp_v4_get_port | -31
tcp_v4_hash | -48
tcp_v4_destroy_sock | -7
tcp_v4_syn_recv_sock | -2
tcp_unhash | -179
5 functions changed, 267 bytes removed, diff: -267
net-2.6/net/ipv6/inet6_hashtables.c:
__inet6_hash | +8
1 function changed, 8 bytes added, diff: +8
net-2.6/net/ipv4/inet_hashtables.c:
inet_unhash | +190
inet_hash | +242
2 functions changed, 432 bytes added, diff: +432
vmlinux:
16 functions changed, 485 bytes added, 492 bytes removed, diff: -7
/home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c:
tcp_v6_get_port | -31
tcp_v6_hash | -7
tcp_v6_syn_recv_sock | -9
3 functions changed, 47 bytes removed, diff: -47
/home/acme/git/net-2.6/net/dccp/proto.c:
dccp_destroy_sock | -7
dccp_unhash | -179
dccp_hash | -49
dccp_set_state | -7
dccp_done | +1
5 functions changed, 1 bytes added, 242 bytes removed, diff: -241
/home/acme/git/net-2.6/net/dccp/ipv4.c:
dccp_v4_get_port | -31
dccp_v4_request_recv_sock | -2
2 functions changed, 33 bytes removed, diff: -33
/home/acme/git/net-2.6/net/dccp/ipv6.c:
dccp_v6_get_port | -31
dccp_v6_hash | -7
dccp_v6_request_recv_sock | +5
3 functions changed, 5 bytes added, 38 bytes removed, diff: -33
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 20:06:04 +08:00
|
|
|
.bind_conflict = inet6_csk_bind_conflict,
|
2006-03-21 14:45:21 +08:00
|
|
|
#ifdef CONFIG_COMPAT
|
2006-03-21 14:48:35 +08:00
|
|
|
.compat_setsockopt = compat_ipv6_setsockopt,
|
|
|
|
.compat_getsockopt = compat_ipv6_getsockopt,
|
2006-03-21 14:45:21 +08:00
|
|
|
#endif
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
2009-09-02 03:25:03 +08:00
|
|
|
static const struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific = {
|
2006-11-15 11:07:45 +08:00
|
|
|
.md5_lookup = tcp_v4_md5_lookup,
|
2008-07-19 15:01:42 +08:00
|
|
|
.calc_md5_hash = tcp_v4_md5_hash_skb,
|
2006-11-15 11:07:45 +08:00
|
|
|
.md5_add = tcp_v6_md5_add_func,
|
|
|
|
.md5_parse = tcp_v6_parse_md5_keys,
|
|
|
|
};
|
2006-11-15 11:53:22 +08:00
|
|
|
#endif
|
2006-11-15 11:07:45 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/* NOTE: A lot of things set to zero explicitly by call to
|
|
|
|
* sk_alloc() so need not be done here.
|
|
|
|
*/
|
|
|
|
static int tcp_v6_init_sock(struct sock *sk)
|
|
|
|
{
|
2005-08-10 15:03:31 +08:00
|
|
|
struct inet_connection_sock *icsk = inet_csk(sk);
|
2005-04-17 06:20:36 +08:00
|
|
|
struct tcp_sock *tp = tcp_sk(sk);
|
|
|
|
|
|
|
|
skb_queue_head_init(&tp->out_of_order_queue);
|
|
|
|
tcp_init_xmit_timers(sk);
|
|
|
|
tcp_prequeue_init(tp);
|
|
|
|
|
2005-08-10 15:03:31 +08:00
|
|
|
icsk->icsk_rto = TCP_TIMEOUT_INIT;
|
2005-04-17 06:20:36 +08:00
|
|
|
tp->mdev = TCP_TIMEOUT_INIT;
|
|
|
|
|
|
|
|
/* So many TCP implementations out there (incorrectly) count the
|
|
|
|
* initial SYN frame in their delayed-ACK and congestion control
|
|
|
|
* algorithms that we must have the following bandaid to talk
|
|
|
|
* efficiently to them. -DaveM
|
|
|
|
*/
|
|
|
|
tp->snd_cwnd = 2;
|
|
|
|
|
|
|
|
/* See draft-stevens-tcpca-spec-01 for discussion of the
|
|
|
|
* initialization of these values.
|
|
|
|
*/
|
2009-09-15 16:30:10 +08:00
|
|
|
tp->snd_ssthresh = TCP_INFINITE_SSTHRESH;
|
2005-04-17 06:20:36 +08:00
|
|
|
tp->snd_cwnd_clamp = ~0;
|
2009-11-10 17:51:18 +08:00
|
|
|
tp->mss_cache = TCP_MSS_DEFAULT;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
tp->reordering = sysctl_tcp_reordering;
|
|
|
|
|
|
|
|
sk->sk_state = TCP_CLOSE;
|
|
|
|
|
2005-12-14 15:15:52 +08:00
|
|
|
icsk->icsk_af_ops = &ipv6_specific;
|
2005-08-10 15:03:31 +08:00
|
|
|
icsk->icsk_ca_ops = &tcp_init_congestion_ops;
|
2005-12-14 15:26:10 +08:00
|
|
|
icsk->icsk_sync_mss = tcp_sync_mss;
|
2005-04-17 06:20:36 +08:00
|
|
|
sk->sk_write_space = sk_stream_write_space;
|
|
|
|
sock_set_flag(sk, SOCK_USE_WRITE_QUEUE);
|
|
|
|
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
tp->af_specific = &tcp_sock_ipv6_specific;
|
|
|
|
#endif
|
|
|
|
|
TCPCT part 1d: define TCP cookie option, extend existing struct's
Data structures are carefully composed to require minimal additions.
For example, the struct tcp_options_received cookie_plus variable fits
between existing 16-bit and 8-bit variables, requiring no additional
space (taking alignment into consideration). There are no additions to
tcp_request_sock, and only 1 pointer in tcp_sock.
This is a significantly revised implementation of an earlier (year-old)
patch that no longer applies cleanly, with permission of the original
author (Adam Langley):
http://thread.gmane.org/gmane.linux.network/102586
The principle difference is using a TCP option to carry the cookie nonce,
instead of a user configured offset in the data. This is more flexible and
less subject to user configuration error. Such a cookie option has been
suggested for many years, and is also useful without SYN data, allowing
several related concepts to use the same extension option.
"Re: SYN floods (was: does history repeat itself?)", September 9, 1996.
http://www.merit.net/mail.archives/nanog/1996-09/msg00235.html
"Re: what a new TCP header might look like", May 12, 1998.
ftp://ftp.isi.edu/end2end/end2end-interest-1998.mail
These functions will also be used in subsequent patches that implement
additional features.
Requires:
TCPCT part 1a: add request_values parameter for sending SYNACK
TCPCT part 1b: generate Responder Cookie secret
TCPCT part 1c: sysctl_tcp_cookie_size, socket option TCP_COOKIE_TRANSACTIONS
Signed-off-by: William.Allen.Simpson@gmail.com
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-03 02:17:05 +08:00
|
|
|
/* TCP Cookie Transactions */
|
|
|
|
if (sysctl_tcp_cookie_size > 0) {
|
|
|
|
/* Default, cookies without s_data_payload. */
|
|
|
|
tp->cookie_values =
|
|
|
|
kzalloc(sizeof(*tp->cookie_values),
|
|
|
|
sk->sk_allocation);
|
|
|
|
if (tp->cookie_values != NULL)
|
|
|
|
kref_init(&tp->cookie_values->kref);
|
|
|
|
}
|
|
|
|
/* Presumed zeroed, in order of appearance:
|
|
|
|
* cookie_in_always, cookie_out_never,
|
|
|
|
* s_data_constant, s_data_in, s_data_out
|
|
|
|
*/
|
2005-04-17 06:20:36 +08:00
|
|
|
sk->sk_sndbuf = sysctl_tcp_wmem[1];
|
|
|
|
sk->sk_rcvbuf = sysctl_tcp_rmem[1];
|
|
|
|
|
2008-12-30 15:04:08 +08:00
|
|
|
local_bh_disable();
|
2008-11-26 13:16:35 +08:00
|
|
|
percpu_counter_inc(&tcp_sockets_allocated);
|
2008-12-30 15:04:08 +08:00
|
|
|
local_bh_enable();
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-06-15 08:04:49 +08:00
|
|
|
static void tcp_v6_destroy_sock(struct sock *sk)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2006-11-15 11:07:45 +08:00
|
|
|
#ifdef CONFIG_TCP_MD5SIG
|
|
|
|
/* Clean up the MD5 key list */
|
|
|
|
if (tcp_sk(sk)->md5sig_info)
|
|
|
|
tcp_v6_clear_md5_list(sk);
|
|
|
|
#endif
|
2005-04-17 06:20:36 +08:00
|
|
|
tcp_v4_destroy_sock(sk);
|
2008-06-15 08:04:49 +08:00
|
|
|
inet6_destroy_sock(sk);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2007-04-21 19:13:44 +08:00
|
|
|
#ifdef CONFIG_PROC_FS
|
2005-04-17 06:20:36 +08:00
|
|
|
/* Proc filesystem TCPv6 sock list dumping. */
|
2007-02-09 22:24:49 +08:00
|
|
|
static void get_openreq6(struct seq_file *seq,
|
2011-10-21 17:22:42 +08:00
|
|
|
const struct sock *sk, struct request_sock *req, int i, int uid)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
int ttd = req->expires - jiffies;
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *src = &inet6_rsk(req)->loc_addr;
|
|
|
|
const struct in6_addr *dest = &inet6_rsk(req)->rmt_addr;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (ttd < 0)
|
|
|
|
ttd = 0;
|
|
|
|
|
|
|
|
seq_printf(seq,
|
|
|
|
"%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
|
net: convert %p usage to %pK
The %pK format specifier is designed to hide exposed kernel pointers,
specifically via /proc interfaces. Exposing these pointers provides an
easy target for kernel write vulnerabilities, since they reveal the
locations of writable structures containing easily triggerable function
pointers. The behavior of %pK depends on the kptr_restrict sysctl.
If kptr_restrict is set to 0, no deviation from the standard %p behavior
occurs. If kptr_restrict is set to 1, the default, if the current user
(intended to be a reader via seq_printf(), etc.) does not have CAP_SYSLOG
(currently in the LSM tree), kernel pointers using %pK are printed as 0's.
If kptr_restrict is set to 2, kernel pointers using %pK are printed as
0's regardless of privileges. Replacing with 0's was chosen over the
default "(null)", which cannot be parsed by userland %p, which expects
"(nil)".
The supporting code for kptr_restrict and %pK are currently in the -mm
tree. This patch converts users of %p in net/ to %pK. Cases of printing
pointers to the syslog are not covered, since this would eliminate useful
information for postmortem debugging and the reading of the syslog is
already optionally protected by the dmesg_restrict sysctl.
Signed-off-by: Dan Rosenberg <drosenberg@vsecurity.com>
Cc: James Morris <jmorris@namei.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Thomas Graf <tgraf@infradead.org>
Cc: Eugene Teo <eugeneteo@kernel.org>
Cc: Kees Cook <kees.cook@canonical.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David S. Miller <davem@davemloft.net>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Eric Paris <eparis@parisplace.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-05-23 20:17:35 +08:00
|
|
|
"%02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %pK\n",
|
2005-04-17 06:20:36 +08:00
|
|
|
i,
|
|
|
|
src->s6_addr32[0], src->s6_addr32[1],
|
|
|
|
src->s6_addr32[2], src->s6_addr32[3],
|
2008-10-20 14:35:58 +08:00
|
|
|
ntohs(inet_rsk(req)->loc_port),
|
2005-04-17 06:20:36 +08:00
|
|
|
dest->s6_addr32[0], dest->s6_addr32[1],
|
|
|
|
dest->s6_addr32[2], dest->s6_addr32[3],
|
[NET] Generalise TCP's struct open_request minisock infrastructure
Kept this first changeset minimal, without changing existing names to
ease peer review.
Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn
has two new members:
->slab, that replaces tcp_openreq_cachep
->obj_size, to inform the size of the openreq descendant for
a specific protocol
The protocol specific fields in struct open_request were moved to a
class hierarchy, with the things that are common to all connection
oriented PF_INET protocols in struct inet_request_sock, the TCP ones
in tcp_request_sock, that is an inet_request_sock, that is an
open_request.
I.e. this uses the same approach used for the struct sock class
hierarchy, with sk_prot indicating if the protocol wants to use the
open_request infrastructure by filling in sk_prot->rsk_prot with an
or_calltable.
Results? Performance is improved and TCP v4 now uses only 64 bytes per
open request minisock, down from 96 without this patch :-)
Next changeset will rename some of the structs, fields and functions
mentioned above, struct or_calltable is way unclear, better name it
struct request_sock_ops, s/struct open_request/struct request_sock/g,
etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-19 13:46:52 +08:00
|
|
|
ntohs(inet_rsk(req)->rmt_port),
|
2005-04-17 06:20:36 +08:00
|
|
|
TCP_SYN_RECV,
|
|
|
|
0,0, /* could print option size, but that is af dependent. */
|
2007-02-09 22:24:49 +08:00
|
|
|
1, /* timers active (only the expire timer) */
|
|
|
|
jiffies_to_clock_t(ttd),
|
2005-04-17 06:20:36 +08:00
|
|
|
req->retrans,
|
|
|
|
uid,
|
2007-02-09 22:24:49 +08:00
|
|
|
0, /* non standard timer */
|
2005-04-17 06:20:36 +08:00
|
|
|
0, /* open_requests have no inode */
|
|
|
|
0, req);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i)
|
|
|
|
{
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *dest, *src;
|
2005-04-17 06:20:36 +08:00
|
|
|
__u16 destp, srcp;
|
|
|
|
int timer_active;
|
|
|
|
unsigned long timer_expires;
|
2011-10-21 17:22:42 +08:00
|
|
|
const struct inet_sock *inet = inet_sk(sp);
|
|
|
|
const struct tcp_sock *tp = tcp_sk(sp);
|
2005-08-10 11:10:42 +08:00
|
|
|
const struct inet_connection_sock *icsk = inet_csk(sp);
|
2011-10-21 17:22:42 +08:00
|
|
|
const struct ipv6_pinfo *np = inet6_sk(sp);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
dest = &np->daddr;
|
|
|
|
src = &np->rcv_saddr;
|
2009-10-15 14:30:45 +08:00
|
|
|
destp = ntohs(inet->inet_dport);
|
|
|
|
srcp = ntohs(inet->inet_sport);
|
2005-08-10 11:10:42 +08:00
|
|
|
|
|
|
|
if (icsk->icsk_pending == ICSK_TIME_RETRANS) {
|
2005-04-17 06:20:36 +08:00
|
|
|
timer_active = 1;
|
2005-08-10 11:10:42 +08:00
|
|
|
timer_expires = icsk->icsk_timeout;
|
|
|
|
} else if (icsk->icsk_pending == ICSK_TIME_PROBE0) {
|
2005-04-17 06:20:36 +08:00
|
|
|
timer_active = 4;
|
2005-08-10 11:10:42 +08:00
|
|
|
timer_expires = icsk->icsk_timeout;
|
2005-04-17 06:20:36 +08:00
|
|
|
} else if (timer_pending(&sp->sk_timer)) {
|
|
|
|
timer_active = 2;
|
|
|
|
timer_expires = sp->sk_timer.expires;
|
|
|
|
} else {
|
|
|
|
timer_active = 0;
|
|
|
|
timer_expires = jiffies;
|
|
|
|
}
|
|
|
|
|
|
|
|
seq_printf(seq,
|
|
|
|
"%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
|
net: convert %p usage to %pK
The %pK format specifier is designed to hide exposed kernel pointers,
specifically via /proc interfaces. Exposing these pointers provides an
easy target for kernel write vulnerabilities, since they reveal the
locations of writable structures containing easily triggerable function
pointers. The behavior of %pK depends on the kptr_restrict sysctl.
If kptr_restrict is set to 0, no deviation from the standard %p behavior
occurs. If kptr_restrict is set to 1, the default, if the current user
(intended to be a reader via seq_printf(), etc.) does not have CAP_SYSLOG
(currently in the LSM tree), kernel pointers using %pK are printed as 0's.
If kptr_restrict is set to 2, kernel pointers using %pK are printed as
0's regardless of privileges. Replacing with 0's was chosen over the
default "(null)", which cannot be parsed by userland %p, which expects
"(nil)".
The supporting code for kptr_restrict and %pK are currently in the -mm
tree. This patch converts users of %p in net/ to %pK. Cases of printing
pointers to the syslog are not covered, since this would eliminate useful
information for postmortem debugging and the reading of the syslog is
already optionally protected by the dmesg_restrict sysctl.
Signed-off-by: Dan Rosenberg <drosenberg@vsecurity.com>
Cc: James Morris <jmorris@namei.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Thomas Graf <tgraf@infradead.org>
Cc: Eugene Teo <eugeneteo@kernel.org>
Cc: Kees Cook <kees.cook@canonical.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David S. Miller <davem@davemloft.net>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Eric Paris <eparis@parisplace.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-05-23 20:17:35 +08:00
|
|
|
"%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %pK %lu %lu %u %u %d\n",
|
2005-04-17 06:20:36 +08:00
|
|
|
i,
|
|
|
|
src->s6_addr32[0], src->s6_addr32[1],
|
|
|
|
src->s6_addr32[2], src->s6_addr32[3], srcp,
|
|
|
|
dest->s6_addr32[0], dest->s6_addr32[1],
|
|
|
|
dest->s6_addr32[2], dest->s6_addr32[3], destp,
|
2007-02-09 22:24:49 +08:00
|
|
|
sp->sk_state,
|
2006-06-28 04:29:00 +08:00
|
|
|
tp->write_seq-tp->snd_una,
|
|
|
|
(sp->sk_state == TCP_LISTEN) ? sp->sk_ack_backlog : (tp->rcv_nxt - tp->copied_seq),
|
2005-04-17 06:20:36 +08:00
|
|
|
timer_active,
|
|
|
|
jiffies_to_clock_t(timer_expires - jiffies),
|
2005-08-10 11:10:42 +08:00
|
|
|
icsk->icsk_retransmits,
|
2005-04-17 06:20:36 +08:00
|
|
|
sock_i_uid(sp),
|
2005-08-10 15:03:31 +08:00
|
|
|
icsk->icsk_probes_out,
|
2005-04-17 06:20:36 +08:00
|
|
|
sock_i_ino(sp),
|
|
|
|
atomic_read(&sp->sk_refcnt), sp,
|
2008-06-28 11:00:19 +08:00
|
|
|
jiffies_to_clock_t(icsk->icsk_rto),
|
|
|
|
jiffies_to_clock_t(icsk->icsk_ack.ato),
|
2005-08-10 11:10:42 +08:00
|
|
|
(icsk->icsk_ack.quick << 1 ) | icsk->icsk_ack.pingpong,
|
2009-09-15 16:30:10 +08:00
|
|
|
tp->snd_cwnd,
|
|
|
|
tcp_in_initial_slowstart(tp) ? -1 : tp->snd_ssthresh
|
2005-04-17 06:20:36 +08:00
|
|
|
);
|
|
|
|
}
|
|
|
|
|
2007-02-09 22:24:49 +08:00
|
|
|
static void get_timewait6_sock(struct seq_file *seq,
|
2005-08-10 11:09:30 +08:00
|
|
|
struct inet_timewait_sock *tw, int i)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2011-04-22 12:53:02 +08:00
|
|
|
const struct in6_addr *dest, *src;
|
2005-04-17 06:20:36 +08:00
|
|
|
__u16 destp, srcp;
|
2011-10-21 17:22:42 +08:00
|
|
|
const struct inet6_timewait_sock *tw6 = inet6_twsk((struct sock *)tw);
|
2005-04-17 06:20:36 +08:00
|
|
|
int ttd = tw->tw_ttd - jiffies;
|
|
|
|
|
|
|
|
if (ttd < 0)
|
|
|
|
ttd = 0;
|
|
|
|
|
2005-12-14 15:23:09 +08:00
|
|
|
dest = &tw6->tw_v6_daddr;
|
|
|
|
src = &tw6->tw_v6_rcv_saddr;
|
2005-04-17 06:20:36 +08:00
|
|
|
destp = ntohs(tw->tw_dport);
|
|
|
|
srcp = ntohs(tw->tw_sport);
|
|
|
|
|
|
|
|
seq_printf(seq,
|
|
|
|
"%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
|
net: convert %p usage to %pK
The %pK format specifier is designed to hide exposed kernel pointers,
specifically via /proc interfaces. Exposing these pointers provides an
easy target for kernel write vulnerabilities, since they reveal the
locations of writable structures containing easily triggerable function
pointers. The behavior of %pK depends on the kptr_restrict sysctl.
If kptr_restrict is set to 0, no deviation from the standard %p behavior
occurs. If kptr_restrict is set to 1, the default, if the current user
(intended to be a reader via seq_printf(), etc.) does not have CAP_SYSLOG
(currently in the LSM tree), kernel pointers using %pK are printed as 0's.
If kptr_restrict is set to 2, kernel pointers using %pK are printed as
0's regardless of privileges. Replacing with 0's was chosen over the
default "(null)", which cannot be parsed by userland %p, which expects
"(nil)".
The supporting code for kptr_restrict and %pK are currently in the -mm
tree. This patch converts users of %p in net/ to %pK. Cases of printing
pointers to the syslog are not covered, since this would eliminate useful
information for postmortem debugging and the reading of the syslog is
already optionally protected by the dmesg_restrict sysctl.
Signed-off-by: Dan Rosenberg <drosenberg@vsecurity.com>
Cc: James Morris <jmorris@namei.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Thomas Graf <tgraf@infradead.org>
Cc: Eugene Teo <eugeneteo@kernel.org>
Cc: Kees Cook <kees.cook@canonical.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David S. Miller <davem@davemloft.net>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Eric Paris <eparis@parisplace.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-05-23 20:17:35 +08:00
|
|
|
"%02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %pK\n",
|
2005-04-17 06:20:36 +08:00
|
|
|
i,
|
|
|
|
src->s6_addr32[0], src->s6_addr32[1],
|
|
|
|
src->s6_addr32[2], src->s6_addr32[3], srcp,
|
|
|
|
dest->s6_addr32[0], dest->s6_addr32[1],
|
|
|
|
dest->s6_addr32[2], dest->s6_addr32[3], destp,
|
|
|
|
tw->tw_substate, 0, 0,
|
|
|
|
3, jiffies_to_clock_t(ttd), 0, 0, 0, 0,
|
|
|
|
atomic_read(&tw->tw_refcnt), tw);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int tcp6_seq_show(struct seq_file *seq, void *v)
|
|
|
|
{
|
|
|
|
struct tcp_iter_state *st;
|
|
|
|
|
|
|
|
if (v == SEQ_START_TOKEN) {
|
|
|
|
seq_puts(seq,
|
|
|
|
" sl "
|
|
|
|
"local_address "
|
|
|
|
"remote_address "
|
|
|
|
"st tx_queue rx_queue tr tm->when retrnsmt"
|
|
|
|
" uid timeout inode\n");
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
st = seq->private;
|
|
|
|
|
|
|
|
switch (st->state) {
|
|
|
|
case TCP_SEQ_STATE_LISTENING:
|
|
|
|
case TCP_SEQ_STATE_ESTABLISHED:
|
|
|
|
get_tcp6_sock(seq, v, st->num);
|
|
|
|
break;
|
|
|
|
case TCP_SEQ_STATE_OPENREQ:
|
|
|
|
get_openreq6(seq, st->syn_wait_sk, v, st->num, st->uid);
|
|
|
|
break;
|
|
|
|
case TCP_SEQ_STATE_TIME_WAIT:
|
|
|
|
get_timewait6_sock(seq, v, st->num);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-10-30 14:46:30 +08:00
|
|
|
static const struct file_operations tcp6_afinfo_seq_fops = {
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.open = tcp_seq_open,
|
|
|
|
.read = seq_read,
|
|
|
|
.llseek = seq_lseek,
|
|
|
|
.release = seq_release_net
|
|
|
|
};
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
static struct tcp_seq_afinfo tcp6_seq_afinfo = {
|
|
|
|
.name = "tcp6",
|
|
|
|
.family = AF_INET6,
|
2011-10-30 14:46:30 +08:00
|
|
|
.seq_fops = &tcp6_afinfo_seq_fops,
|
2008-04-14 13:12:13 +08:00
|
|
|
.seq_ops = {
|
|
|
|
.show = tcp6_seq_show,
|
|
|
|
},
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2010-01-17 11:35:32 +08:00
|
|
|
int __net_init tcp6_proc_init(struct net *net)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2008-03-21 19:14:45 +08:00
|
|
|
return tcp_proc_register(net, &tcp6_seq_afinfo);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2008-03-21 19:14:45 +08:00
|
|
|
void tcp6_proc_exit(struct net *net)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2008-03-21 19:14:45 +08:00
|
|
|
tcp_proc_unregister(net, &tcp6_seq_afinfo);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
struct proto tcpv6_prot = {
|
|
|
|
.name = "TCPv6",
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.close = tcp_close,
|
|
|
|
.connect = tcp_v6_connect,
|
|
|
|
.disconnect = tcp_disconnect,
|
2005-08-10 11:10:42 +08:00
|
|
|
.accept = inet_csk_accept,
|
2005-04-17 06:20:36 +08:00
|
|
|
.ioctl = tcp_ioctl,
|
|
|
|
.init = tcp_v6_init_sock,
|
|
|
|
.destroy = tcp_v6_destroy_sock,
|
|
|
|
.shutdown = tcp_shutdown,
|
|
|
|
.setsockopt = tcp_setsockopt,
|
|
|
|
.getsockopt = tcp_getsockopt,
|
|
|
|
.recvmsg = tcp_recvmsg,
|
2010-07-11 04:41:55 +08:00
|
|
|
.sendmsg = tcp_sendmsg,
|
|
|
|
.sendpage = tcp_sendpage,
|
2005-04-17 06:20:36 +08:00
|
|
|
.backlog_rcv = tcp_v6_do_rcv,
|
|
|
|
.hash = tcp_v6_hash,
|
[SOCK] proto: Add hashinfo member to struct proto
This way we can remove TCP and DCCP specific versions of
sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port
sk->sk_prot->hash: inet_hash is directly used, only v6 need
a specific version to deal with mapped sockets
sk->sk_prot->unhash: both v4 and v6 use inet_hash directly
struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so
that inet_csk_get_port can find the per family routine.
Now only the lookup routines receive as a parameter a struct inet_hashtable.
With this we further reuse code, reducing the difference among INET transport
protocols.
Eventually work has to be done on UDP and SCTP to make them share this
infrastructure and get as a bonus inet_diag interfaces so that iproute can be
used with these protocols.
net-2.6/net/ipv4/inet_hashtables.c:
struct proto | +8
struct inet_connection_sock_af_ops | +8
2 structs changed
__inet_hash_nolisten | +18
__inet_hash | -210
inet_put_port | +8
inet_bind_bucket_create | +1
__inet_hash_connect | -8
5 functions changed, 27 bytes added, 218 bytes removed, diff: -191
net-2.6/net/core/sock.c:
proto_seq_show | +3
1 function changed, 3 bytes added, diff: +3
net-2.6/net/ipv4/inet_connection_sock.c:
inet_csk_get_port | +15
1 function changed, 15 bytes added, diff: +15
net-2.6/net/ipv4/tcp.c:
tcp_set_state | -7
1 function changed, 7 bytes removed, diff: -7
net-2.6/net/ipv4/tcp_ipv4.c:
tcp_v4_get_port | -31
tcp_v4_hash | -48
tcp_v4_destroy_sock | -7
tcp_v4_syn_recv_sock | -2
tcp_unhash | -179
5 functions changed, 267 bytes removed, diff: -267
net-2.6/net/ipv6/inet6_hashtables.c:
__inet6_hash | +8
1 function changed, 8 bytes added, diff: +8
net-2.6/net/ipv4/inet_hashtables.c:
inet_unhash | +190
inet_hash | +242
2 functions changed, 432 bytes added, diff: +432
vmlinux:
16 functions changed, 485 bytes added, 492 bytes removed, diff: -7
/home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c:
tcp_v6_get_port | -31
tcp_v6_hash | -7
tcp_v6_syn_recv_sock | -9
3 functions changed, 47 bytes removed, diff: -47
/home/acme/git/net-2.6/net/dccp/proto.c:
dccp_destroy_sock | -7
dccp_unhash | -179
dccp_hash | -49
dccp_set_state | -7
dccp_done | +1
5 functions changed, 1 bytes added, 242 bytes removed, diff: -241
/home/acme/git/net-2.6/net/dccp/ipv4.c:
dccp_v4_get_port | -31
dccp_v4_request_recv_sock | -2
2 functions changed, 33 bytes removed, diff: -33
/home/acme/git/net-2.6/net/dccp/ipv6.c:
dccp_v6_get_port | -31
dccp_v6_hash | -7
dccp_v6_request_recv_sock | +5
3 functions changed, 5 bytes added, 38 bytes removed, diff: -33
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 20:06:04 +08:00
|
|
|
.unhash = inet_unhash,
|
|
|
|
.get_port = inet_csk_get_port,
|
2005-04-17 06:20:36 +08:00
|
|
|
.enter_memory_pressure = tcp_enter_memory_pressure,
|
|
|
|
.sockets_allocated = &tcp_sockets_allocated,
|
|
|
|
.memory_allocated = &tcp_memory_allocated,
|
|
|
|
.memory_pressure = &tcp_memory_pressure,
|
2005-08-10 11:11:41 +08:00
|
|
|
.orphan_count = &tcp_orphan_count,
|
2005-04-17 06:20:36 +08:00
|
|
|
.sysctl_mem = sysctl_tcp_mem,
|
|
|
|
.sysctl_wmem = sysctl_tcp_wmem,
|
|
|
|
.sysctl_rmem = sysctl_tcp_rmem,
|
|
|
|
.max_header = MAX_TCP_HEADER,
|
|
|
|
.obj_size = sizeof(struct tcp6_sock),
|
2008-11-17 11:40:17 +08:00
|
|
|
.slab_flags = SLAB_DESTROY_BY_RCU,
|
2005-12-14 15:25:19 +08:00
|
|
|
.twsk_prot = &tcp6_timewait_sock_ops,
|
2005-06-19 13:47:21 +08:00
|
|
|
.rsk_prot = &tcp6_request_sock_ops,
|
2008-03-23 07:50:58 +08:00
|
|
|
.h.hashinfo = &tcp_hashinfo,
|
2010-07-11 04:41:55 +08:00
|
|
|
.no_autobind = true,
|
2006-03-21 14:48:35 +08:00
|
|
|
#ifdef CONFIG_COMPAT
|
|
|
|
.compat_setsockopt = compat_tcp_setsockopt,
|
|
|
|
.compat_getsockopt = compat_tcp_getsockopt,
|
|
|
|
#endif
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2009-09-14 20:22:28 +08:00
|
|
|
static const struct inet6_protocol tcpv6_protocol = {
|
2005-04-17 06:20:36 +08:00
|
|
|
.handler = tcp_v6_rcv,
|
|
|
|
.err_handler = tcp_v6_err,
|
2006-07-09 04:34:56 +08:00
|
|
|
.gso_send_check = tcp_v6_gso_send_check,
|
2006-07-01 04:36:15 +08:00
|
|
|
.gso_segment = tcp_tso_segment,
|
2009-01-09 02:41:23 +08:00
|
|
|
.gro_receive = tcp6_gro_receive,
|
|
|
|
.gro_complete = tcp6_gro_complete,
|
2005-04-17 06:20:36 +08:00
|
|
|
.flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct inet_protosw tcpv6_protosw = {
|
|
|
|
.type = SOCK_STREAM,
|
|
|
|
.protocol = IPPROTO_TCP,
|
|
|
|
.prot = &tcpv6_prot,
|
|
|
|
.ops = &inet6_stream_ops,
|
|
|
|
.no_check = 0,
|
2005-12-14 15:26:10 +08:00
|
|
|
.flags = INET_PROTOSW_PERMANENT |
|
|
|
|
INET_PROTOSW_ICSK,
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2010-01-17 11:35:32 +08:00
|
|
|
static int __net_init tcpv6_net_init(struct net *net)
|
2008-03-08 03:16:02 +08:00
|
|
|
{
|
2008-04-04 05:28:30 +08:00
|
|
|
return inet_ctl_sock_create(&net->ipv6.tcp_sk, PF_INET6,
|
|
|
|
SOCK_RAW, IPPROTO_TCP, net);
|
2008-03-08 03:16:02 +08:00
|
|
|
}
|
|
|
|
|
2010-01-17 11:35:32 +08:00
|
|
|
static void __net_exit tcpv6_net_exit(struct net *net)
|
2008-03-08 03:16:02 +08:00
|
|
|
{
|
2008-04-04 05:28:30 +08:00
|
|
|
inet_ctl_sock_destroy(net->ipv6.tcp_sk);
|
2009-12-03 10:29:09 +08:00
|
|
|
}
|
|
|
|
|
2010-01-17 11:35:32 +08:00
|
|
|
static void __net_exit tcpv6_net_exit_batch(struct list_head *net_exit_list)
|
2009-12-03 10:29:09 +08:00
|
|
|
{
|
|
|
|
inet_twsk_purge(&tcp_hashinfo, &tcp_death_row, AF_INET6);
|
2008-03-08 03:16:02 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct pernet_operations tcpv6_net_ops = {
|
2009-12-03 10:29:09 +08:00
|
|
|
.init = tcpv6_net_init,
|
|
|
|
.exit = tcpv6_net_exit,
|
|
|
|
.exit_batch = tcpv6_net_exit_batch,
|
2008-03-08 03:16:02 +08:00
|
|
|
};
|
|
|
|
|
2007-12-11 18:25:35 +08:00
|
|
|
int __init tcpv6_init(void)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2007-12-11 18:25:35 +08:00
|
|
|
int ret;
|
|
|
|
|
|
|
|
ret = inet6_add_protocol(&tcpv6_protocol, IPPROTO_TCP);
|
|
|
|
if (ret)
|
|
|
|
goto out;
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/* register inet6 protocol */
|
2007-12-11 18:25:35 +08:00
|
|
|
ret = inet6_register_protosw(&tcpv6_protosw);
|
|
|
|
if (ret)
|
|
|
|
goto out_tcpv6_protocol;
|
|
|
|
|
2008-03-08 03:16:02 +08:00
|
|
|
ret = register_pernet_subsys(&tcpv6_net_ops);
|
2007-12-11 18:25:35 +08:00
|
|
|
if (ret)
|
|
|
|
goto out_tcpv6_protosw;
|
|
|
|
out:
|
|
|
|
return ret;
|
2006-01-12 07:53:04 +08:00
|
|
|
|
2007-12-11 18:25:35 +08:00
|
|
|
out_tcpv6_protocol:
|
|
|
|
inet6_del_protocol(&tcpv6_protocol, IPPROTO_TCP);
|
|
|
|
out_tcpv6_protosw:
|
|
|
|
inet6_unregister_protosw(&tcpv6_protosw);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2007-12-13 21:34:58 +08:00
|
|
|
void tcpv6_exit(void)
|
2007-12-11 18:25:35 +08:00
|
|
|
{
|
2008-03-08 03:16:02 +08:00
|
|
|
unregister_pernet_subsys(&tcpv6_net_ops);
|
2007-12-11 18:25:35 +08:00
|
|
|
inet6_unregister_protosw(&tcpv6_protosw);
|
|
|
|
inet6_del_protocol(&tcpv6_protocol, IPPROTO_TCP);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|