2005-04-17 05:20:36 +07:00
|
|
|
/*
|
|
|
|
* INET An implementation of the TCP/IP protocol suite for the LINUX
|
|
|
|
* operating system. INET is implemented using the BSD Socket
|
|
|
|
* interface as the means of communication with the user level.
|
|
|
|
*
|
|
|
|
* RAW - implementation of IP "raw" sockets.
|
|
|
|
*
|
2005-05-06 06:16:16 +07:00
|
|
|
* Authors: Ross Biro
|
2005-04-17 05:20:36 +07:00
|
|
|
* Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
|
|
|
|
*
|
|
|
|
* Fixes:
|
|
|
|
* Alan Cox : verify_area() fixed up
|
|
|
|
* Alan Cox : ICMP error handling
|
|
|
|
* Alan Cox : EMSGSIZE if you send too big a packet
|
|
|
|
* Alan Cox : Now uses generic datagrams and shared
|
|
|
|
* skbuff library. No more peek crashes,
|
|
|
|
* no more backlogs
|
|
|
|
* Alan Cox : Checks sk->broadcast.
|
|
|
|
* Alan Cox : Uses skb_free_datagram/skb_copy_datagram
|
|
|
|
* Alan Cox : Raw passes ip options too
|
|
|
|
* Alan Cox : Setsocketopt added
|
|
|
|
* Alan Cox : Fixed error return for broadcasts
|
|
|
|
* Alan Cox : Removed wake_up calls
|
|
|
|
* Alan Cox : Use ttl/tos
|
|
|
|
* Alan Cox : Cleaned up old debugging
|
|
|
|
* Alan Cox : Use new kernel side addresses
|
|
|
|
* Arnt Gulbrandsen : Fixed MSG_DONTROUTE in raw sockets.
|
|
|
|
* Alan Cox : BSD style RAW socket demultiplexing.
|
|
|
|
* Alan Cox : Beginnings of mrouted support.
|
|
|
|
* Alan Cox : Added IP_HDRINCL option.
|
|
|
|
* Alan Cox : Skip broadcast check if BSDism set.
|
|
|
|
* David S. Miller : New socket lookup architecture.
|
|
|
|
*
|
|
|
|
* 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.
|
|
|
|
*/
|
2006-09-23 04:00:29 +07:00
|
|
|
|
2006-01-19 08:44:07 +07:00
|
|
|
#include <linux/types.h>
|
2011-07-27 06:09:06 +07:00
|
|
|
#include <linux/atomic.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
#include <asm/byteorder.h>
|
|
|
|
#include <asm/current.h>
|
|
|
|
#include <asm/uaccess.h>
|
|
|
|
#include <asm/ioctls.h>
|
|
|
|
#include <linux/stddef.h>
|
|
|
|
#include <linux/slab.h>
|
|
|
|
#include <linux/errno.h>
|
|
|
|
#include <linux/aio.h>
|
|
|
|
#include <linux/kernel.h>
|
2011-07-15 22:47:34 +07:00
|
|
|
#include <linux/export.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
#include <linux/spinlock.h>
|
|
|
|
#include <linux/sockios.h>
|
|
|
|
#include <linux/socket.h>
|
|
|
|
#include <linux/in.h>
|
|
|
|
#include <linux/mroute.h>
|
|
|
|
#include <linux/netdevice.h>
|
|
|
|
#include <linux/in_route.h>
|
|
|
|
#include <linux/route.h>
|
|
|
|
#include <linux/skbuff.h>
|
2007-09-12 17:01:34 +07:00
|
|
|
#include <net/net_namespace.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
#include <net/dst.h>
|
|
|
|
#include <net/sock.h>
|
|
|
|
#include <linux/ip.h>
|
|
|
|
#include <linux/net.h>
|
|
|
|
#include <net/ip.h>
|
|
|
|
#include <net/icmp.h>
|
|
|
|
#include <net/udp.h>
|
|
|
|
#include <net/raw.h>
|
|
|
|
#include <net/snmp.h>
|
2005-08-10 10:08:28 +07:00
|
|
|
#include <net/tcp_states.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
#include <net/inet_common.h>
|
|
|
|
#include <net/checksum.h>
|
|
|
|
#include <net/xfrm.h>
|
|
|
|
#include <linux/rtnetlink.h>
|
|
|
|
#include <linux/proc_fs.h>
|
|
|
|
#include <linux/seq_file.h>
|
|
|
|
#include <linux/netfilter.h>
|
|
|
|
#include <linux/netfilter_ipv4.h>
|
2011-01-29 23:15:56 +07:00
|
|
|
#include <linux/compat.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2007-11-20 13:36:45 +07:00
|
|
|
static struct raw_hashinfo raw_v4_hashinfo = {
|
2008-03-18 14:59:23 +07:00
|
|
|
.lock = __RW_LOCK_UNLOCKED(raw_v4_hashinfo.lock),
|
2007-11-20 13:36:45 +07:00
|
|
|
};
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-03-23 06:56:51 +07:00
|
|
|
void raw_hash_sk(struct sock *sk)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-03-23 06:56:51 +07:00
|
|
|
struct raw_hashinfo *h = sk->sk_prot->h.raw_hash;
|
2007-11-20 13:37:24 +07:00
|
|
|
struct hlist_head *head;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-10-15 13:30:45 +07:00
|
|
|
head = &h->ht[inet_sk(sk)->inet_num & (RAW_HTABLE_SIZE - 1)];
|
2007-11-20 13:37:24 +07:00
|
|
|
|
|
|
|
write_lock_bh(&h->lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
sk_add_node(sk, head);
|
2008-04-01 09:41:46 +07:00
|
|
|
sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
|
2007-11-20 13:37:24 +07:00
|
|
|
write_unlock_bh(&h->lock);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(raw_hash_sk);
|
|
|
|
|
2008-03-23 06:56:51 +07:00
|
|
|
void raw_unhash_sk(struct sock *sk)
|
2007-11-20 13:37:58 +07:00
|
|
|
{
|
2008-03-23 06:56:51 +07:00
|
|
|
struct raw_hashinfo *h = sk->sk_prot->h.raw_hash;
|
|
|
|
|
2007-11-20 13:37:58 +07:00
|
|
|
write_lock_bh(&h->lock);
|
|
|
|
if (sk_del_node_init(sk))
|
2008-04-01 09:41:46 +07:00
|
|
|
sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
|
2007-11-20 13:37:58 +07:00
|
|
|
write_unlock_bh(&h->lock);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(raw_unhash_sk);
|
|
|
|
|
2008-01-14 20:35:31 +07:00
|
|
|
static struct sock *__raw_v4_lookup(struct net *net, struct sock *sk,
|
|
|
|
unsigned short num, __be32 raddr, __be32 laddr, int dif)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:06:00 +07:00
|
|
|
sk_for_each_from(sk) {
|
2005-04-17 05:20:36 +07:00
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
|
|
|
|
2009-10-15 13:30:45 +07:00
|
|
|
if (net_eq(sock_net(sk), net) && inet->inet_num == num &&
|
|
|
|
!(inet->inet_daddr && inet->inet_daddr != raddr) &&
|
|
|
|
!(inet->inet_rcv_saddr && inet->inet_rcv_saddr != laddr) &&
|
2005-04-17 05:20:36 +07:00
|
|
|
!(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif))
|
|
|
|
goto found; /* gotcha */
|
|
|
|
}
|
|
|
|
sk = NULL;
|
|
|
|
found:
|
|
|
|
return sk;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 0 - deliver
|
|
|
|
* 1 - block
|
|
|
|
*/
|
2012-09-22 07:08:29 +07:00
|
|
|
static int icmp_filter(const struct sock *sk, const struct sk_buff *skb)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2012-09-22 07:08:29 +07:00
|
|
|
struct icmphdr _hdr;
|
|
|
|
const struct icmphdr *hdr;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2012-09-22 07:08:29 +07:00
|
|
|
hdr = skb_header_pointer(skb, skb_transport_offset(skb),
|
|
|
|
sizeof(_hdr), &_hdr);
|
|
|
|
if (!hdr)
|
2005-04-17 05:20:36 +07:00
|
|
|
return 1;
|
|
|
|
|
2012-09-22 07:08:29 +07:00
|
|
|
if (hdr->type < 32) {
|
2005-04-17 05:20:36 +07:00
|
|
|
__u32 data = raw_sk(sk)->filter.data;
|
|
|
|
|
2012-09-22 07:08:29 +07:00
|
|
|
return ((1U << hdr->type) & data) != 0;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Do not block unknown ICMP types */
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* IP input processing comes here for RAW socket delivery.
|
|
|
|
* Caller owns SKB, so we must make clones.
|
|
|
|
*
|
|
|
|
* RFC 1122: SHOULD pass TOS value up to the transport layer.
|
|
|
|
* -> It does. And not only TOS, but all IP header.
|
|
|
|
*/
|
2011-04-22 11:53:02 +07:00
|
|
|
static int raw_v4_input(struct sk_buff *skb, const struct iphdr *iph, int hash)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
struct sock *sk;
|
|
|
|
struct hlist_head *head;
|
2005-08-10 09:45:02 +07:00
|
|
|
int delivered = 0;
|
2008-01-14 20:35:31 +07:00
|
|
|
struct net *net;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2007-11-20 13:36:45 +07:00
|
|
|
read_lock(&raw_v4_hashinfo.lock);
|
|
|
|
head = &raw_v4_hashinfo.ht[hash];
|
2005-04-17 05:20:36 +07:00
|
|
|
if (hlist_empty(head))
|
|
|
|
goto out;
|
2008-01-14 20:35:31 +07:00
|
|
|
|
2008-03-25 19:47:49 +07:00
|
|
|
net = dev_net(skb->dev);
|
2008-01-14 20:35:31 +07:00
|
|
|
sk = __raw_v4_lookup(net, __sk_head(head), iph->protocol,
|
2005-04-17 05:20:36 +07:00
|
|
|
iph->saddr, iph->daddr,
|
|
|
|
skb->dev->ifindex);
|
|
|
|
|
|
|
|
while (sk) {
|
2005-08-10 09:45:02 +07:00
|
|
|
delivered = 1;
|
2005-04-17 05:20:36 +07:00
|
|
|
if (iph->protocol != IPPROTO_ICMP || !icmp_filter(sk, skb)) {
|
|
|
|
struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC);
|
|
|
|
|
|
|
|
/* Not releasing hash table! */
|
|
|
|
if (clone)
|
|
|
|
raw_rcv(sk, clone);
|
|
|
|
}
|
2008-01-14 20:35:31 +07:00
|
|
|
sk = __raw_v4_lookup(net, sk_next(sk), iph->protocol,
|
2005-04-17 05:20:36 +07:00
|
|
|
iph->saddr, iph->daddr,
|
|
|
|
skb->dev->ifindex);
|
|
|
|
}
|
|
|
|
out:
|
2007-11-20 13:36:45 +07:00
|
|
|
read_unlock(&raw_v4_hashinfo.lock);
|
2005-08-10 09:45:02 +07:00
|
|
|
return delivered;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2007-11-20 13:35:07 +07:00
|
|
|
int raw_local_deliver(struct sk_buff *skb, int protocol)
|
|
|
|
{
|
|
|
|
int hash;
|
|
|
|
struct sock *raw_sk;
|
|
|
|
|
2007-11-20 13:36:45 +07:00
|
|
|
hash = protocol & (RAW_HTABLE_SIZE - 1);
|
|
|
|
raw_sk = sk_head(&raw_v4_hashinfo.ht[hash]);
|
2007-11-20 13:35:07 +07:00
|
|
|
|
|
|
|
/* If there maybe a raw socket we must check - if not we
|
|
|
|
* don't care less
|
|
|
|
*/
|
|
|
|
if (raw_sk && !raw_v4_input(skb, ip_hdr(skb), hash))
|
|
|
|
raw_sk = NULL;
|
|
|
|
|
|
|
|
return raw_sk != NULL;
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
static void raw_err(struct sock *sk, struct sk_buff *skb, u32 info)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
2007-03-14 00:43:18 +07:00
|
|
|
const int type = icmp_hdr(skb)->type;
|
|
|
|
const int code = icmp_hdr(skb)->code;
|
2005-04-17 05:20:36 +07:00
|
|
|
int err = 0;
|
|
|
|
int harderr = 0;
|
|
|
|
|
2012-06-15 12:21:46 +07:00
|
|
|
if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED)
|
|
|
|
ipv4_sk_update_pmtu(skb, sk, info);
|
2013-09-20 17:21:25 +07:00
|
|
|
else if (type == ICMP_REDIRECT) {
|
2012-07-12 11:27:49 +07:00
|
|
|
ipv4_sk_redirect(skb, sk);
|
2013-09-20 17:21:25 +07:00
|
|
|
return;
|
|
|
|
}
|
2012-06-15 12:21:46 +07:00
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
/* Report error on raw socket, if:
|
|
|
|
1. User requested ip_recverr.
|
|
|
|
2. Socket is connected (otherwise the error indication
|
|
|
|
is useless without ip_recverr and error is hard.
|
|
|
|
*/
|
|
|
|
if (!inet->recverr && sk->sk_state != TCP_ESTABLISHED)
|
|
|
|
return;
|
|
|
|
|
|
|
|
switch (type) {
|
|
|
|
default:
|
|
|
|
case ICMP_TIME_EXCEEDED:
|
|
|
|
err = EHOSTUNREACH;
|
|
|
|
break;
|
|
|
|
case ICMP_SOURCE_QUENCH:
|
|
|
|
return;
|
|
|
|
case ICMP_PARAMETERPROB:
|
|
|
|
err = EPROTO;
|
|
|
|
harderr = 1;
|
|
|
|
break;
|
|
|
|
case ICMP_DEST_UNREACH:
|
|
|
|
err = EHOSTUNREACH;
|
|
|
|
if (code > NR_ICMP_UNREACH)
|
|
|
|
break;
|
|
|
|
err = icmp_err_convert[code].errno;
|
|
|
|
harderr = icmp_err_convert[code].fatal;
|
|
|
|
if (code == ICMP_FRAG_NEEDED) {
|
|
|
|
harderr = inet->pmtudisc != IP_PMTUDISC_DONT;
|
|
|
|
err = EMSGSIZE;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (inet->recverr) {
|
2011-04-22 11:53:02 +07:00
|
|
|
const struct iphdr *iph = (const struct iphdr *)skb->data;
|
2005-04-17 05:20:36 +07:00
|
|
|
u8 *payload = skb->data + (iph->ihl << 2);
|
|
|
|
|
|
|
|
if (inet->hdrincl)
|
|
|
|
payload = skb->data;
|
|
|
|
ip_icmp_error(sk, skb, err, 0, info, payload);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (inet->recverr || harderr) {
|
|
|
|
sk->sk_err = err;
|
|
|
|
sk->sk_error_report(sk);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2007-11-20 13:35:07 +07:00
|
|
|
void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info)
|
|
|
|
{
|
|
|
|
int hash;
|
|
|
|
struct sock *raw_sk;
|
2011-04-22 11:53:02 +07:00
|
|
|
const struct iphdr *iph;
|
2008-01-14 20:35:31 +07:00
|
|
|
struct net *net;
|
2007-11-20 13:35:07 +07:00
|
|
|
|
2007-11-20 13:36:45 +07:00
|
|
|
hash = protocol & (RAW_HTABLE_SIZE - 1);
|
2007-11-20 13:35:07 +07:00
|
|
|
|
2007-11-20 13:36:45 +07:00
|
|
|
read_lock(&raw_v4_hashinfo.lock);
|
|
|
|
raw_sk = sk_head(&raw_v4_hashinfo.ht[hash]);
|
2007-11-20 13:35:07 +07:00
|
|
|
if (raw_sk != NULL) {
|
2011-04-22 11:53:02 +07:00
|
|
|
iph = (const struct iphdr *)skb->data;
|
2008-03-25 19:47:49 +07:00
|
|
|
net = dev_net(skb->dev);
|
2008-01-14 20:35:31 +07:00
|
|
|
|
|
|
|
while ((raw_sk = __raw_v4_lookup(net, raw_sk, protocol,
|
|
|
|
iph->daddr, iph->saddr,
|
2007-11-20 13:35:07 +07:00
|
|
|
skb->dev->ifindex)) != NULL) {
|
|
|
|
raw_err(raw_sk, skb, info);
|
|
|
|
raw_sk = sk_next(raw_sk);
|
2011-04-22 11:53:02 +07:00
|
|
|
iph = (const struct iphdr *)skb->data;
|
2007-11-20 13:35:07 +07:00
|
|
|
}
|
|
|
|
}
|
2007-11-20 13:36:45 +07:00
|
|
|
read_unlock(&raw_v4_hashinfo.lock);
|
2007-11-20 13:35:07 +07:00
|
|
|
}
|
|
|
|
|
2012-04-15 08:34:41 +07:00
|
|
|
static int raw_rcv_skb(struct sock *sk, struct sk_buff *skb)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
/* Charge it to the socket. */
|
2007-02-09 21:24:47 +07:00
|
|
|
|
2013-10-07 23:01:40 +07:00
|
|
|
ipv4_pktinfo_prepare(sk, skb);
|
ipv4: PKTINFO doesnt need dst reference
Le lundi 07 novembre 2011 à 15:33 +0100, Eric Dumazet a écrit :
> At least, in recent kernels we dont change dst->refcnt in forwarding
> patch (usinf NOREF skb->dst)
>
> One particular point is the atomic_inc(dst->refcnt) we have to perform
> when queuing an UDP packet if socket asked PKTINFO stuff (for example a
> typical DNS server has to setup this option)
>
> I have one patch somewhere that stores the information in skb->cb[] and
> avoid the atomic_{inc|dec}(dst->refcnt).
>
OK I found it, I did some extra tests and believe its ready.
[PATCH net-next] ipv4: IP_PKTINFO doesnt need dst reference
When a socket uses IP_PKTINFO notifications, we currently force a dst
reference for each received skb. Reader has to access dst to get needed
information (rt_iif & rt_spec_dst) and must release dst reference.
We also forced a dst reference if skb was put in socket backlog, even
without IP_PKTINFO handling. This happens under stress/load.
We can instead store the needed information in skb->cb[], so that only
softirq handler really access dst, improving cache hit ratios.
This removes two atomic operations per packet, and false sharing as
well.
On a benchmark using a mono threaded receiver (doing only recvmsg()
calls), I can reach 720.000 pps instead of 570.000 pps.
IP_PKTINFO is typically used by DNS servers, and any multihomed aware
UDP application.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-09 14:24:35 +07:00
|
|
|
if (sock_queue_rcv_skb(sk, skb) < 0) {
|
2005-04-17 05:20:36 +07:00
|
|
|
kfree_skb(skb);
|
|
|
|
return NET_RX_DROP;
|
|
|
|
}
|
|
|
|
|
|
|
|
return NET_RX_SUCCESS;
|
|
|
|
}
|
|
|
|
|
|
|
|
int raw_rcv(struct sock *sk, struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) {
|
2007-11-14 11:30:01 +07:00
|
|
|
atomic_inc(&sk->sk_drops);
|
2005-04-17 05:20:36 +07:00
|
|
|
kfree_skb(skb);
|
|
|
|
return NET_RX_DROP;
|
|
|
|
}
|
2006-01-07 14:06:10 +07:00
|
|
|
nf_reset(skb);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2007-04-11 10:50:43 +07:00
|
|
|
skb_push(skb, skb->data - skb_network_header(skb));
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
raw_rcv_skb(sk, skb);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-05-09 07:12:19 +07:00
|
|
|
static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
|
|
|
|
void *from, size_t length,
|
|
|
|
struct rtable **rtp,
|
|
|
|
unsigned int flags)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
2008-07-15 13:00:43 +07:00
|
|
|
struct net *net = sock_net(sk);
|
2005-04-17 05:20:36 +07:00
|
|
|
struct iphdr *iph;
|
|
|
|
struct sk_buff *skb;
|
2008-01-06 14:14:44 +07:00
|
|
|
unsigned int iphlen;
|
2005-04-17 05:20:36 +07:00
|
|
|
int err;
|
2010-06-04 05:23:57 +07:00
|
|
|
struct rtable *rt = *rtp;
|
2011-11-18 09:20:04 +07:00
|
|
|
int hlen, tlen;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-06-11 13:31:35 +07:00
|
|
|
if (length > rt->dst.dev->mtu) {
|
2011-05-09 07:12:19 +07:00
|
|
|
ip_local_error(sk, EMSGSIZE, fl4->daddr, inet->inet_dport,
|
2010-06-11 13:31:35 +07:00
|
|
|
rt->dst.dev->mtu);
|
2005-04-17 05:20:36 +07:00
|
|
|
return -EMSGSIZE;
|
|
|
|
}
|
|
|
|
if (flags&MSG_PROBE)
|
|
|
|
goto out;
|
|
|
|
|
2011-11-18 09:20:04 +07:00
|
|
|
hlen = LL_RESERVED_SPACE(rt->dst.dev);
|
|
|
|
tlen = rt->dst.dev->needed_tailroom;
|
2008-05-13 10:48:31 +07:00
|
|
|
skb = sock_alloc_send_skb(sk,
|
2011-11-18 09:20:04 +07:00
|
|
|
length + hlen + tlen + 15,
|
2008-05-13 10:48:31 +07:00
|
|
|
flags & MSG_DONTWAIT, &err);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (skb == NULL)
|
2007-02-09 21:24:47 +07:00
|
|
|
goto error;
|
2011-11-18 09:20:04 +07:00
|
|
|
skb_reserve(skb, hlen);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
skb->priority = sk->sk_priority;
|
2008-01-31 10:08:16 +07:00
|
|
|
skb->mark = sk->sk_mark;
|
2010-06-11 13:31:35 +07:00
|
|
|
skb_dst_set(skb, &rt->dst);
|
2010-06-04 05:23:57 +07:00
|
|
|
*rtp = NULL;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2007-03-11 04:40:59 +07:00
|
|
|
skb_reset_network_header(skb);
|
2007-04-21 12:47:35 +07:00
|
|
|
iph = ip_hdr(skb);
|
2007-03-11 04:40:59 +07:00
|
|
|
skb_put(skb, length);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
skb->ip_summed = CHECKSUM_NONE;
|
|
|
|
|
2007-04-11 11:21:55 +07:00
|
|
|
skb->transport_header = skb->network_header;
|
AF_RAW: Augment raw_send_hdrinc to expand skb to fit iphdr->ihl (v2)
Augment raw_send_hdrinc to correct for incorrect ip header length values
A series of oopses was reported to me recently. Apparently when using AF_RAW
sockets to send data to peers that were reachable via ipsec encapsulation,
people could panic or BUG halt their systems.
I've tracked the problem down to user space sending an invalid ip header over an
AF_RAW socket with IP_HDRINCL set to 1.
Basically what happens is that userspace sends down an ip frame that includes
only the header (no data), but sets the ip header ihl value to a large number,
one that is larger than the total amount of data passed to the sendmsg call. In
raw_send_hdrincl, we allocate an skb based on the size of the data in the msghdr
that was passed in, but assume the data is all valid. Later during ipsec
encapsulation, xfrm4_tranport_output moves the entire frame back in the skbuff
to provide headroom for the ipsec headers. During this operation, the
skb->transport_header is repointed to a spot computed by
skb->network_header + the ip header length (ihl). Since so little data was
passed in relative to the value of ihl provided by the raw socket, we point
transport header to an unknown location, resulting in various crashes.
This fix for this is pretty straightforward, simply validate the value of of
iph->ihl when sending over a raw socket. If (iph->ihl*4U) > user data buffer
size, drop the frame and return -EINVAL. I just confirmed this fixes the
reported crashes.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-28 15:59:47 +07:00
|
|
|
err = -EFAULT;
|
|
|
|
if (memcpy_fromiovecend((void *)iph, from, 0, length))
|
|
|
|
goto error_free;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-01-06 14:14:44 +07:00
|
|
|
iphlen = iph->ihl * 4;
|
AF_RAW: Augment raw_send_hdrinc to expand skb to fit iphdr->ihl (v2)
Augment raw_send_hdrinc to correct for incorrect ip header length values
A series of oopses was reported to me recently. Apparently when using AF_RAW
sockets to send data to peers that were reachable via ipsec encapsulation,
people could panic or BUG halt their systems.
I've tracked the problem down to user space sending an invalid ip header over an
AF_RAW socket with IP_HDRINCL set to 1.
Basically what happens is that userspace sends down an ip frame that includes
only the header (no data), but sets the ip header ihl value to a large number,
one that is larger than the total amount of data passed to the sendmsg call. In
raw_send_hdrincl, we allocate an skb based on the size of the data in the msghdr
that was passed in, but assume the data is all valid. Later during ipsec
encapsulation, xfrm4_tranport_output moves the entire frame back in the skbuff
to provide headroom for the ipsec headers. During this operation, the
skb->transport_header is repointed to a spot computed by
skb->network_header + the ip header length (ihl). Since so little data was
passed in relative to the value of ihl provided by the raw socket, we point
transport header to an unknown location, resulting in various crashes.
This fix for this is pretty straightforward, simply validate the value of of
iph->ihl when sending over a raw socket. If (iph->ihl*4U) > user data buffer
size, drop the frame and return -EINVAL. I just confirmed this fixes the
reported crashes.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-28 15:59:47 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* We don't want to modify the ip header, but we do need to
|
|
|
|
* be sure that it won't cause problems later along the network
|
|
|
|
* stack. Specifically we want to make sure that iph->ihl is a
|
|
|
|
* sane value. If ihl points beyond the length of the buffer passed
|
|
|
|
* in, reject the frame as invalid
|
|
|
|
*/
|
|
|
|
err = -EINVAL;
|
|
|
|
if (iphlen > length)
|
|
|
|
goto error_free;
|
|
|
|
|
|
|
|
if (iphlen >= sizeof(*iph)) {
|
2005-04-17 05:20:36 +07:00
|
|
|
if (!iph->saddr)
|
2011-05-09 07:12:19 +07:00
|
|
|
iph->saddr = fl4->saddr;
|
2005-04-17 05:20:36 +07:00
|
|
|
iph->check = 0;
|
|
|
|
iph->tot_len = htons(length);
|
|
|
|
if (!iph->id)
|
2013-09-19 05:29:53 +07:00
|
|
|
ip_select_ident(skb, &rt->dst, NULL);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
|
|
|
|
}
|
2007-09-17 23:57:33 +07:00
|
|
|
if (iph->protocol == IPPROTO_ICMP)
|
2008-07-15 13:00:43 +07:00
|
|
|
icmp_out_count(net, ((struct icmphdr *)
|
2007-09-17 23:57:33 +07:00
|
|
|
skb_transport_header(skb))->type);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-03-23 10:07:29 +07:00
|
|
|
err = NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_OUT, skb, NULL,
|
2010-06-11 13:31:35 +07:00
|
|
|
rt->dst.dev, dst_output);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (err > 0)
|
ip: Report qdisc packet drops
Christoph Lameter pointed out that packet drops at qdisc level where not
accounted in SNMP counters. Only if application sets IP_RECVERR, drops
are reported to user (-ENOBUFS errors) and SNMP counters updated.
IP_RECVERR is used to enable extended reliable error message passing,
but these are not needed to update system wide SNMP stats.
This patch changes things a bit to allow SNMP counters to be updated,
regardless of IP_RECVERR being set or not on the socket.
Example after an UDP tx flood
# netstat -s
...
IP:
1487048 outgoing packets dropped
...
Udp:
...
SndbufErrors: 1487048
send() syscalls, do however still return an OK status, to not
break applications.
Note : send() manual page explicitly says for -ENOBUFS error :
"The output queue for a network interface was full.
This generally indicates that the interface has stopped sending,
but may be caused by transient congestion.
(Normally, this does not occur in Linux. Packets are just silently
dropped when a device queue overflows.) "
This is not true for IP_RECVERR enabled sockets : a send() syscall
that hit a qdisc drop returns an ENOBUFS error.
Many thanks to Christoph, David, and last but not least, Alexey !
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 08:05:33 +07:00
|
|
|
err = net_xmit_errno(err);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (err)
|
|
|
|
goto error;
|
|
|
|
out:
|
|
|
|
return 0;
|
|
|
|
|
AF_RAW: Augment raw_send_hdrinc to expand skb to fit iphdr->ihl (v2)
Augment raw_send_hdrinc to correct for incorrect ip header length values
A series of oopses was reported to me recently. Apparently when using AF_RAW
sockets to send data to peers that were reachable via ipsec encapsulation,
people could panic or BUG halt their systems.
I've tracked the problem down to user space sending an invalid ip header over an
AF_RAW socket with IP_HDRINCL set to 1.
Basically what happens is that userspace sends down an ip frame that includes
only the header (no data), but sets the ip header ihl value to a large number,
one that is larger than the total amount of data passed to the sendmsg call. In
raw_send_hdrincl, we allocate an skb based on the size of the data in the msghdr
that was passed in, but assume the data is all valid. Later during ipsec
encapsulation, xfrm4_tranport_output moves the entire frame back in the skbuff
to provide headroom for the ipsec headers. During this operation, the
skb->transport_header is repointed to a spot computed by
skb->network_header + the ip header length (ihl). Since so little data was
passed in relative to the value of ihl provided by the raw socket, we point
transport header to an unknown location, resulting in various crashes.
This fix for this is pretty straightforward, simply validate the value of of
iph->ihl when sending over a raw socket. If (iph->ihl*4U) > user data buffer
size, drop the frame and return -EINVAL. I just confirmed this fixes the
reported crashes.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-28 15:59:47 +07:00
|
|
|
error_free:
|
2005-04-17 05:20:36 +07:00
|
|
|
kfree_skb(skb);
|
|
|
|
error:
|
2008-07-17 10:19:49 +07:00
|
|
|
IP_INC_STATS(net, IPSTATS_MIB_OUTDISCARDS);
|
ip: Report qdisc packet drops
Christoph Lameter pointed out that packet drops at qdisc level where not
accounted in SNMP counters. Only if application sets IP_RECVERR, drops
are reported to user (-ENOBUFS errors) and SNMP counters updated.
IP_RECVERR is used to enable extended reliable error message passing,
but these are not needed to update system wide SNMP stats.
This patch changes things a bit to allow SNMP counters to be updated,
regardless of IP_RECVERR being set or not on the socket.
Example after an UDP tx flood
# netstat -s
...
IP:
1487048 outgoing packets dropped
...
Udp:
...
SndbufErrors: 1487048
send() syscalls, do however still return an OK status, to not
break applications.
Note : send() manual page explicitly says for -ENOBUFS error :
"The output queue for a network interface was full.
This generally indicates that the interface has stopped sending,
but may be caused by transient congestion.
(Normally, this does not occur in Linux. Packets are just silently
dropped when a device queue overflows.) "
This is not true for IP_RECVERR enabled sockets : a send() syscall
that hit a qdisc drop returns an ENOBUFS error.
Many thanks to Christoph, David, and last but not least, Alexey !
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 08:05:33 +07:00
|
|
|
if (err == -ENOBUFS && !inet->recverr)
|
|
|
|
err = 0;
|
2007-02-09 21:24:47 +07:00
|
|
|
return err;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2011-03-12 13:12:47 +07:00
|
|
|
static int raw_probe_proto_opt(struct flowi4 *fl4, struct msghdr *msg)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
struct iovec *iov;
|
|
|
|
u8 __user *type = NULL;
|
|
|
|
u8 __user *code = NULL;
|
|
|
|
int probed = 0;
|
2005-06-19 13:00:15 +07:00
|
|
|
unsigned int i;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
if (!msg->msg_iov)
|
2006-10-31 06:06:12 +07:00
|
|
|
return 0;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
for (i = 0; i < msg->msg_iovlen; i++) {
|
|
|
|
iov = &msg->msg_iov[i];
|
|
|
|
if (!iov)
|
|
|
|
continue;
|
|
|
|
|
2011-03-12 13:12:47 +07:00
|
|
|
switch (fl4->flowi4_proto) {
|
2005-04-17 05:20:36 +07:00
|
|
|
case IPPROTO_ICMP:
|
|
|
|
/* check if one-byte field is readable or not. */
|
|
|
|
if (iov->iov_base && iov->iov_len < 1)
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (!type) {
|
|
|
|
type = iov->iov_base;
|
|
|
|
/* check if code field is readable or not. */
|
|
|
|
if (iov->iov_len > 1)
|
|
|
|
code = type + 1;
|
|
|
|
} else if (!code)
|
|
|
|
code = iov->iov_base;
|
|
|
|
|
|
|
|
if (type && code) {
|
2011-03-12 15:00:33 +07:00
|
|
|
if (get_user(fl4->fl4_icmp_type, type) ||
|
|
|
|
get_user(fl4->fl4_icmp_code, code))
|
2006-10-31 06:06:12 +07:00
|
|
|
return -EFAULT;
|
2005-04-17 05:20:36 +07:00
|
|
|
probed = 1;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
probed = 1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (probed)
|
|
|
|
break;
|
|
|
|
}
|
2006-10-31 06:06:12 +07:00
|
|
|
return 0;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
|
|
|
|
size_t len)
|
|
|
|
{
|
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
|
|
|
struct ipcm_cookie ipc;
|
|
|
|
struct rtable *rt = NULL;
|
2011-05-09 07:12:19 +07:00
|
|
|
struct flowi4 fl4;
|
2005-04-17 05:20:36 +07:00
|
|
|
int free = 0;
|
2006-09-28 08:28:07 +07:00
|
|
|
__be32 daddr;
|
2006-09-28 08:28:28 +07:00
|
|
|
__be32 saddr;
|
2005-04-17 05:20:36 +07:00
|
|
|
u8 tos;
|
|
|
|
int err;
|
2011-04-21 16:45:37 +07:00
|
|
|
struct ip_options_data opt_copy;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
err = -EMSGSIZE;
|
2005-06-19 13:00:00 +07:00
|
|
|
if (len > 0xFFFF)
|
2005-04-17 05:20:36 +07:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Check the flags.
|
|
|
|
*/
|
|
|
|
|
|
|
|
err = -EOPNOTSUPP;
|
|
|
|
if (msg->msg_flags & MSG_OOB) /* Mirror BSD error message */
|
|
|
|
goto out; /* compatibility */
|
2007-02-09 21:24:47 +07:00
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
/*
|
2007-02-09 21:24:47 +07:00
|
|
|
* Get and verify the address.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
|
|
|
|
|
|
|
if (msg->msg_namelen) {
|
2014-01-18 04:53:15 +07:00
|
|
|
DECLARE_SOCKADDR(struct sockaddr_in *, usin, msg->msg_name);
|
2005-04-17 05:20:36 +07:00
|
|
|
err = -EINVAL;
|
|
|
|
if (msg->msg_namelen < sizeof(*usin))
|
|
|
|
goto out;
|
|
|
|
if (usin->sin_family != AF_INET) {
|
2012-03-12 01:36:11 +07:00
|
|
|
pr_info_once("%s: %s forgot to set AF_INET. Fix it!\n",
|
|
|
|
__func__, current->comm);
|
2005-04-17 05:20:36 +07:00
|
|
|
err = -EAFNOSUPPORT;
|
|
|
|
if (usin->sin_family)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
daddr = usin->sin_addr.s_addr;
|
|
|
|
/* ANK: I did not forget to get protocol from port field.
|
|
|
|
* I just do not know, who uses this weirdness.
|
|
|
|
* IP_HDRINCL is much more convenient.
|
|
|
|
*/
|
|
|
|
} else {
|
|
|
|
err = -EDESTADDRREQ;
|
2007-02-09 21:24:47 +07:00
|
|
|
if (sk->sk_state != TCP_ESTABLISHED)
|
2005-04-17 05:20:36 +07:00
|
|
|
goto out;
|
2009-10-15 13:30:45 +07:00
|
|
|
daddr = inet->inet_daddr;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2009-10-15 13:30:45 +07:00
|
|
|
ipc.addr = inet->inet_saddr;
|
2005-04-17 05:20:36 +07:00
|
|
|
ipc.opt = NULL;
|
2010-08-17 15:59:14 +07:00
|
|
|
ipc.tx_flags = 0;
|
2013-09-24 20:43:09 +07:00
|
|
|
ipc.ttl = 0;
|
|
|
|
ipc.tos = -1;
|
2005-04-17 05:20:36 +07:00
|
|
|
ipc.oif = sk->sk_bound_dev_if;
|
|
|
|
|
|
|
|
if (msg->msg_controllen) {
|
2014-02-19 03:38:08 +07:00
|
|
|
err = ip_cmsg_send(sock_net(sk), msg, &ipc, false);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
if (ipc.opt)
|
|
|
|
free = 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
saddr = ipc.addr;
|
|
|
|
ipc.addr = daddr;
|
|
|
|
|
2011-04-21 16:45:37 +07:00
|
|
|
if (!ipc.opt) {
|
|
|
|
struct ip_options_rcu *inet_opt;
|
|
|
|
|
|
|
|
rcu_read_lock();
|
|
|
|
inet_opt = rcu_dereference(inet->inet_opt);
|
|
|
|
if (inet_opt) {
|
|
|
|
memcpy(&opt_copy, inet_opt,
|
|
|
|
sizeof(*inet_opt) + inet_opt->opt.optlen);
|
|
|
|
ipc.opt = &opt_copy.opt;
|
|
|
|
}
|
|
|
|
rcu_read_unlock();
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
if (ipc.opt) {
|
|
|
|
err = -EINVAL;
|
|
|
|
/* Linux does not mangle headers on raw sockets,
|
|
|
|
* so that IP options + IP_HDRINCL is non-sense.
|
|
|
|
*/
|
|
|
|
if (inet->hdrincl)
|
|
|
|
goto done;
|
2011-04-21 16:45:37 +07:00
|
|
|
if (ipc.opt->opt.srr) {
|
2005-04-17 05:20:36 +07:00
|
|
|
if (!daddr)
|
|
|
|
goto done;
|
2011-04-21 16:45:37 +07:00
|
|
|
daddr = ipc.opt->opt.faddr;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
}
|
2013-09-24 20:43:09 +07:00
|
|
|
tos = get_rtconn_flags(&ipc, sk);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (msg->msg_flags & MSG_DONTROUTE)
|
|
|
|
tos |= RTO_ONLINK;
|
|
|
|
|
2007-12-17 04:45:43 +07:00
|
|
|
if (ipv4_is_multicast(daddr)) {
|
2005-04-17 05:20:36 +07:00
|
|
|
if (!ipc.oif)
|
|
|
|
ipc.oif = inet->mc_index;
|
|
|
|
if (!saddr)
|
|
|
|
saddr = inet->mc_addr;
|
2012-02-08 16:11:07 +07:00
|
|
|
} else if (!ipc.oif)
|
|
|
|
ipc.oif = inet->uc_index;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2011-05-09 07:12:19 +07:00
|
|
|
flowi4_init_output(&fl4, ipc.oif, sk->sk_mark, tos,
|
|
|
|
RT_SCOPE_UNIVERSE,
|
|
|
|
inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol,
|
2013-08-28 13:04:14 +07:00
|
|
|
inet_sk_flowi_flags(sk) |
|
2013-08-28 01:02:15 +07:00
|
|
|
(inet->hdrincl ? FLOWI_FLAG_KNOWN_NH : 0),
|
2011-08-07 16:16:09 +07:00
|
|
|
daddr, saddr, 0, 0);
|
2011-03-31 18:53:51 +07:00
|
|
|
|
2011-05-09 07:12:19 +07:00
|
|
|
if (!inet->hdrincl) {
|
|
|
|
err = raw_probe_proto_opt(&fl4, msg);
|
|
|
|
if (err)
|
2011-03-03 05:31:35 +07:00
|
|
|
goto done;
|
2011-05-09 07:12:19 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
security_sk_classify_flow(sk, flowi4_to_flowi(&fl4));
|
|
|
|
rt = ip_route_output_flow(sock_net(sk), &fl4, sk);
|
|
|
|
if (IS_ERR(rt)) {
|
|
|
|
err = PTR_ERR(rt);
|
|
|
|
rt = NULL;
|
|
|
|
goto done;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
err = -EACCES;
|
|
|
|
if (rt->rt_flags & RTCF_BROADCAST && !sock_flag(sk, SOCK_BROADCAST))
|
|
|
|
goto done;
|
|
|
|
|
|
|
|
if (msg->msg_flags & MSG_CONFIRM)
|
|
|
|
goto do_confirm;
|
|
|
|
back_from_confirm:
|
|
|
|
|
|
|
|
if (inet->hdrincl)
|
2011-05-09 07:12:19 +07:00
|
|
|
err = raw_send_hdrinc(sk, &fl4, msg->msg_iov, len,
|
|
|
|
&rt, msg->msg_flags);
|
2007-02-09 21:24:47 +07:00
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
else {
|
|
|
|
if (!ipc.addr)
|
2011-05-09 07:12:19 +07:00
|
|
|
ipc.addr = fl4.daddr;
|
2005-04-17 05:20:36 +07:00
|
|
|
lock_sock(sk);
|
2011-05-09 07:24:10 +07:00
|
|
|
err = ip_append_data(sk, &fl4, ip_generic_getfrag,
|
2011-05-09 07:12:19 +07:00
|
|
|
msg->msg_iov, len, 0,
|
|
|
|
&ipc, &rt, msg->msg_flags);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (err)
|
|
|
|
ip_flush_pending_frames(sk);
|
ip: Report qdisc packet drops
Christoph Lameter pointed out that packet drops at qdisc level where not
accounted in SNMP counters. Only if application sets IP_RECVERR, drops
are reported to user (-ENOBUFS errors) and SNMP counters updated.
IP_RECVERR is used to enable extended reliable error message passing,
but these are not needed to update system wide SNMP stats.
This patch changes things a bit to allow SNMP counters to be updated,
regardless of IP_RECVERR being set or not on the socket.
Example after an UDP tx flood
# netstat -s
...
IP:
1487048 outgoing packets dropped
...
Udp:
...
SndbufErrors: 1487048
send() syscalls, do however still return an OK status, to not
break applications.
Note : send() manual page explicitly says for -ENOBUFS error :
"The output queue for a network interface was full.
This generally indicates that the interface has stopped sending,
but may be caused by transient congestion.
(Normally, this does not occur in Linux. Packets are just silently
dropped when a device queue overflows.) "
This is not true for IP_RECVERR enabled sockets : a send() syscall
that hit a qdisc drop returns an ENOBUFS error.
Many thanks to Christoph, David, and last but not least, Alexey !
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 08:05:33 +07:00
|
|
|
else if (!(msg->msg_flags & MSG_MORE)) {
|
2011-05-09 07:12:19 +07:00
|
|
|
err = ip_push_pending_frames(sk, &fl4);
|
ip: Report qdisc packet drops
Christoph Lameter pointed out that packet drops at qdisc level where not
accounted in SNMP counters. Only if application sets IP_RECVERR, drops
are reported to user (-ENOBUFS errors) and SNMP counters updated.
IP_RECVERR is used to enable extended reliable error message passing,
but these are not needed to update system wide SNMP stats.
This patch changes things a bit to allow SNMP counters to be updated,
regardless of IP_RECVERR being set or not on the socket.
Example after an UDP tx flood
# netstat -s
...
IP:
1487048 outgoing packets dropped
...
Udp:
...
SndbufErrors: 1487048
send() syscalls, do however still return an OK status, to not
break applications.
Note : send() manual page explicitly says for -ENOBUFS error :
"The output queue for a network interface was full.
This generally indicates that the interface has stopped sending,
but may be caused by transient congestion.
(Normally, this does not occur in Linux. Packets are just silently
dropped when a device queue overflows.) "
This is not true for IP_RECVERR enabled sockets : a send() syscall
that hit a qdisc drop returns an ENOBUFS error.
Many thanks to Christoph, David, and last but not least, Alexey !
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 08:05:33 +07:00
|
|
|
if (err == -ENOBUFS && !inet->recverr)
|
|
|
|
err = 0;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
release_sock(sk);
|
|
|
|
}
|
|
|
|
done:
|
|
|
|
if (free)
|
|
|
|
kfree(ipc.opt);
|
|
|
|
ip_rt_put(rt);
|
|
|
|
|
2005-06-19 12:59:45 +07:00
|
|
|
out:
|
|
|
|
if (err < 0)
|
|
|
|
return err;
|
|
|
|
return len;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
do_confirm:
|
2010-06-11 13:31:35 +07:00
|
|
|
dst_confirm(&rt->dst);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (!(msg->msg_flags & MSG_PROBE) || len)
|
|
|
|
goto back_from_confirm;
|
|
|
|
err = 0;
|
|
|
|
goto done;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void raw_close(struct sock *sk, long timeout)
|
|
|
|
{
|
2007-02-09 21:24:47 +07:00
|
|
|
/*
|
2011-03-31 08:57:33 +07:00
|
|
|
* Raw sockets may have direct kernel references. Kill them.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
|
|
|
ip_ra_control(sk, 0, NULL);
|
|
|
|
|
|
|
|
sk_common_release(sk);
|
|
|
|
}
|
|
|
|
|
2008-06-15 07:04:49 +07:00
|
|
|
static void raw_destroy(struct sock *sk)
|
raw: Raw socket leak.
The program below just leaks the raw kernel socket
int main() {
int fd = socket(PF_INET, SOCK_RAW, IPPROTO_UDP);
struct sockaddr_in addr;
memset(&addr, 0, sizeof(addr));
inet_aton("127.0.0.1", &addr.sin_addr);
addr.sin_family = AF_INET;
addr.sin_port = htons(2048);
sendto(fd, "a", 1, MSG_MORE, &addr, sizeof(addr));
return 0;
}
Corked packet is allocated via sock_wmalloc which holds the owner socket,
so one should uncork it and flush all pending data on close. Do this in the
same way as in UDP.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-06-05 05:16:12 +07:00
|
|
|
{
|
|
|
|
lock_sock(sk);
|
|
|
|
ip_flush_pending_frames(sk);
|
|
|
|
release_sock(sk);
|
|
|
|
}
|
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
/* This gets rid of all the nasties in af_inet. -DaveM */
|
|
|
|
static int raw_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
|
|
|
|
{
|
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
|
|
|
struct sockaddr_in *addr = (struct sockaddr_in *) uaddr;
|
|
|
|
int ret = -EINVAL;
|
|
|
|
int chk_addr_ret;
|
|
|
|
|
|
|
|
if (sk->sk_state != TCP_CLOSE || addr_len < sizeof(struct sockaddr_in))
|
|
|
|
goto out;
|
2008-03-26 00:26:21 +07:00
|
|
|
chk_addr_ret = inet_addr_type(sock_net(sk), addr->sin_addr.s_addr);
|
2005-04-17 05:20:36 +07:00
|
|
|
ret = -EADDRNOTAVAIL;
|
|
|
|
if (addr->sin_addr.s_addr && chk_addr_ret != RTN_LOCAL &&
|
|
|
|
chk_addr_ret != RTN_MULTICAST && chk_addr_ret != RTN_BROADCAST)
|
|
|
|
goto out;
|
2009-10-15 13:30:45 +07:00
|
|
|
inet->inet_rcv_saddr = inet->inet_saddr = addr->sin_addr.s_addr;
|
2005-04-17 05:20:36 +07:00
|
|
|
if (chk_addr_ret == RTN_MULTICAST || chk_addr_ret == RTN_BROADCAST)
|
2009-10-15 13:30:45 +07:00
|
|
|
inet->inet_saddr = 0; /* Use device */
|
2005-04-17 05:20:36 +07:00
|
|
|
sk_dst_reset(sk);
|
|
|
|
ret = 0;
|
|
|
|
out: return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This should be easy, if there is something there
|
|
|
|
* we return it, otherwise we block.
|
|
|
|
*/
|
|
|
|
|
|
|
|
static int raw_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
|
|
|
|
size_t len, int noblock, int flags, int *addr_len)
|
|
|
|
{
|
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
|
|
|
size_t copied = 0;
|
|
|
|
int err = -EOPNOTSUPP;
|
2014-01-18 04:53:15 +07:00
|
|
|
DECLARE_SOCKADDR(struct sockaddr_in *, sin, msg->msg_name);
|
2005-04-17 05:20:36 +07:00
|
|
|
struct sk_buff *skb;
|
|
|
|
|
|
|
|
if (flags & MSG_OOB)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
if (flags & MSG_ERRQUEUE) {
|
2013-11-23 06:46:12 +07:00
|
|
|
err = ip_recv_error(sk, msg, len, addr_len);
|
2005-04-17 05:20:36 +07:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
skb = skb_recv_datagram(sk, flags, noblock, &err);
|
|
|
|
if (!skb)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
copied = skb->len;
|
|
|
|
if (len < copied) {
|
|
|
|
msg->msg_flags |= MSG_TRUNC;
|
|
|
|
copied = len;
|
|
|
|
}
|
|
|
|
|
|
|
|
err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
|
|
|
|
if (err)
|
|
|
|
goto done;
|
|
|
|
|
net: Generalize socket rx gap / receive queue overflow cmsg
Create a new socket level option to report number of queue overflows
Recently I augmented the AF_PACKET protocol to report the number of frames lost
on the socket receive queue between any two enqueued frames. This value was
exported via a SOL_PACKET level cmsg. AFter I completed that work it was
requested that this feature be generalized so that any datagram oriented socket
could make use of this option. As such I've created this patch, It creates a
new SOL_SOCKET level option called SO_RXQ_OVFL, which when enabled exports a
SOL_SOCKET level cmsg that reports the nubmer of times the sk_receive_queue
overflowed between any two given frames. It also augments the AF_PACKET
protocol to take advantage of this new feature (as it previously did not touch
sk->sk_drops, which this patch uses to record the overflow count). Tested
successfully by me.
Notes:
1) Unlike my previous patch, this patch simply records the sk_drops value, which
is not a number of drops between packets, but rather a total number of drops.
Deltas must be computed in user space.
2) While this patch currently works with datagram oriented protocols, it will
also be accepted by non-datagram oriented protocols. I'm not sure if thats
agreeable to everyone, but my argument in favor of doing so is that, for those
protocols which aren't applicable to this option, sk_drops will always be zero,
and reporting no drops on a receive queue that isn't used for those
non-participating protocols seems reasonable to me. This also saves us having
to code in a per-protocol opt in mechanism.
3) This applies cleanly to net-next assuming that commit
977750076d98c7ff6cbda51858bb5a5894a9d9ab (my af packet cmsg patch) is reverted
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-13 03:26:31 +07:00
|
|
|
sock_recv_ts_and_drops(msg, sk, skb);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
/* Copy the address. */
|
|
|
|
if (sin) {
|
|
|
|
sin->sin_family = AF_INET;
|
2007-04-21 12:47:35 +07:00
|
|
|
sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
|
2006-07-26 07:05:35 +07:00
|
|
|
sin->sin_port = 0;
|
2005-04-17 05:20:36 +07:00
|
|
|
memset(&sin->sin_zero, 0, sizeof(sin->sin_zero));
|
2013-11-18 10:20:45 +07:00
|
|
|
*addr_len = sizeof(*sin);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
if (inet->cmsg_flags)
|
|
|
|
ip_cmsg_recv(msg, skb);
|
|
|
|
if (flags & MSG_TRUNC)
|
|
|
|
copied = skb->len;
|
|
|
|
done:
|
|
|
|
skb_free_datagram(sk, skb);
|
2005-06-19 12:59:45 +07:00
|
|
|
out:
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
return copied;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
static int raw_init(struct sock *sk)
|
|
|
|
{
|
|
|
|
struct raw_sock *rp = raw_sk(sk);
|
|
|
|
|
2009-10-15 13:30:45 +07:00
|
|
|
if (inet_sk(sk)->inet_num == IPPROTO_ICMP)
|
2005-04-17 05:20:36 +07:00
|
|
|
memset(&rp->filter, 0, sizeof(rp->filter));
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int raw_seticmpfilter(struct sock *sk, char __user *optval, int optlen)
|
|
|
|
{
|
|
|
|
if (optlen > sizeof(struct icmp_filter))
|
|
|
|
optlen = sizeof(struct icmp_filter);
|
|
|
|
if (copy_from_user(&raw_sk(sk)->filter, optval, optlen))
|
|
|
|
return -EFAULT;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int raw_geticmpfilter(struct sock *sk, char __user *optval, int __user *optlen)
|
|
|
|
{
|
|
|
|
int len, ret = -EFAULT;
|
|
|
|
|
|
|
|
if (get_user(len, optlen))
|
|
|
|
goto out;
|
|
|
|
ret = -EINVAL;
|
|
|
|
if (len < 0)
|
|
|
|
goto out;
|
|
|
|
if (len > sizeof(struct icmp_filter))
|
|
|
|
len = sizeof(struct icmp_filter);
|
|
|
|
ret = -EFAULT;
|
|
|
|
if (put_user(len, optlen) ||
|
|
|
|
copy_to_user(optval, &raw_sk(sk)->filter, len))
|
|
|
|
goto out;
|
|
|
|
ret = 0;
|
|
|
|
out: return ret;
|
|
|
|
}
|
|
|
|
|
2006-03-21 13:45:21 +07:00
|
|
|
static int do_raw_setsockopt(struct sock *sk, int level, int optname,
|
2009-10-01 06:12:20 +07:00
|
|
|
char __user *optval, unsigned int optlen)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
if (optname == ICMP_FILTER) {
|
2009-10-15 13:30:45 +07:00
|
|
|
if (inet_sk(sk)->inet_num != IPPROTO_ICMP)
|
2005-04-17 05:20:36 +07:00
|
|
|
return -EOPNOTSUPP;
|
|
|
|
else
|
|
|
|
return raw_seticmpfilter(sk, optval, optlen);
|
|
|
|
}
|
|
|
|
return -ENOPROTOOPT;
|
|
|
|
}
|
|
|
|
|
2006-03-21 13:45:21 +07:00
|
|
|
static int raw_setsockopt(struct sock *sk, int level, int optname,
|
2009-10-01 06:12:20 +07:00
|
|
|
char __user *optval, unsigned int optlen)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
if (level != SOL_RAW)
|
2006-03-21 13:45:21 +07:00
|
|
|
return ip_setsockopt(sk, level, optname, optval, optlen);
|
|
|
|
return do_raw_setsockopt(sk, level, optname, optval, optlen);
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2006-03-21 13:45:21 +07:00
|
|
|
#ifdef CONFIG_COMPAT
|
|
|
|
static int compat_raw_setsockopt(struct sock *sk, int level, int optname,
|
2009-10-01 06:12:20 +07:00
|
|
|
char __user *optval, unsigned int optlen)
|
2006-03-21 13:45:21 +07:00
|
|
|
{
|
|
|
|
if (level != SOL_RAW)
|
2006-03-21 13:48:35 +07:00
|
|
|
return compat_ip_setsockopt(sk, level, optname, optval, optlen);
|
2006-03-21 13:45:21 +07:00
|
|
|
return do_raw_setsockopt(sk, level, optname, optval, optlen);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
static int do_raw_getsockopt(struct sock *sk, int level, int optname,
|
|
|
|
char __user *optval, int __user *optlen)
|
|
|
|
{
|
2005-04-17 05:20:36 +07:00
|
|
|
if (optname == ICMP_FILTER) {
|
2009-10-15 13:30:45 +07:00
|
|
|
if (inet_sk(sk)->inet_num != IPPROTO_ICMP)
|
2005-04-17 05:20:36 +07:00
|
|
|
return -EOPNOTSUPP;
|
|
|
|
else
|
|
|
|
return raw_geticmpfilter(sk, optval, optlen);
|
|
|
|
}
|
|
|
|
return -ENOPROTOOPT;
|
|
|
|
}
|
|
|
|
|
2006-03-21 13:45:21 +07:00
|
|
|
static int raw_getsockopt(struct sock *sk, int level, int optname,
|
|
|
|
char __user *optval, int __user *optlen)
|
|
|
|
{
|
|
|
|
if (level != SOL_RAW)
|
|
|
|
return ip_getsockopt(sk, level, optname, optval, optlen);
|
|
|
|
return do_raw_getsockopt(sk, level, optname, optval, optlen);
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef CONFIG_COMPAT
|
|
|
|
static int compat_raw_getsockopt(struct sock *sk, int level, int optname,
|
2006-03-21 13:48:35 +07:00
|
|
|
char __user *optval, int __user *optlen)
|
2006-03-21 13:45:21 +07:00
|
|
|
{
|
|
|
|
if (level != SOL_RAW)
|
2006-03-21 13:48:35 +07:00
|
|
|
return compat_ip_getsockopt(sk, level, optname, optval, optlen);
|
2006-03-21 13:45:21 +07:00
|
|
|
return do_raw_getsockopt(sk, level, optname, optval, optlen);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
static int raw_ioctl(struct sock *sk, int cmd, unsigned long arg)
|
|
|
|
{
|
|
|
|
switch (cmd) {
|
2011-07-01 16:43:07 +07:00
|
|
|
case SIOCOUTQ: {
|
|
|
|
int amount = sk_wmem_alloc_get(sk);
|
2009-06-18 09:05:41 +07:00
|
|
|
|
2011-07-01 16:43:07 +07:00
|
|
|
return put_user(amount, (int __user *)arg);
|
|
|
|
}
|
|
|
|
case SIOCINQ: {
|
|
|
|
struct sk_buff *skb;
|
|
|
|
int amount = 0;
|
|
|
|
|
|
|
|
spin_lock_bh(&sk->sk_receive_queue.lock);
|
|
|
|
skb = skb_peek(&sk->sk_receive_queue);
|
|
|
|
if (skb != NULL)
|
|
|
|
amount = skb->len;
|
|
|
|
spin_unlock_bh(&sk->sk_receive_queue.lock);
|
|
|
|
return put_user(amount, (int __user *)arg);
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2011-07-01 16:43:07 +07:00
|
|
|
default:
|
2005-04-17 05:20:36 +07:00
|
|
|
#ifdef CONFIG_IP_MROUTE
|
2011-07-01 16:43:07 +07:00
|
|
|
return ipmr_ioctl(sk, cmd, (void __user *)arg);
|
2005-04-17 05:20:36 +07:00
|
|
|
#else
|
2011-07-01 16:43:07 +07:00
|
|
|
return -ENOIOCTLCMD;
|
2005-04-17 05:20:36 +07:00
|
|
|
#endif
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2011-01-29 23:15:56 +07:00
|
|
|
#ifdef CONFIG_COMPAT
|
|
|
|
static int compat_raw_ioctl(struct sock *sk, unsigned int cmd, unsigned long arg)
|
|
|
|
{
|
|
|
|
switch (cmd) {
|
|
|
|
case SIOCOUTQ:
|
|
|
|
case SIOCINQ:
|
|
|
|
return -ENOIOCTLCMD;
|
|
|
|
default:
|
|
|
|
#ifdef CONFIG_IP_MROUTE
|
|
|
|
return ipmr_compat_ioctl(sk, cmd, compat_ptr(arg));
|
|
|
|
#else
|
|
|
|
return -ENOIOCTLCMD;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
struct proto raw_prot = {
|
2006-03-21 13:48:35 +07:00
|
|
|
.name = "RAW",
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.close = raw_close,
|
raw: Raw socket leak.
The program below just leaks the raw kernel socket
int main() {
int fd = socket(PF_INET, SOCK_RAW, IPPROTO_UDP);
struct sockaddr_in addr;
memset(&addr, 0, sizeof(addr));
inet_aton("127.0.0.1", &addr.sin_addr);
addr.sin_family = AF_INET;
addr.sin_port = htons(2048);
sendto(fd, "a", 1, MSG_MORE, &addr, sizeof(addr));
return 0;
}
Corked packet is allocated via sock_wmalloc which holds the owner socket,
so one should uncork it and flush all pending data on close. Do this in the
same way as in UDP.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-06-05 05:16:12 +07:00
|
|
|
.destroy = raw_destroy,
|
2006-03-21 13:48:35 +07:00
|
|
|
.connect = ip4_datagram_connect,
|
|
|
|
.disconnect = udp_disconnect,
|
|
|
|
.ioctl = raw_ioctl,
|
|
|
|
.init = raw_init,
|
|
|
|
.setsockopt = raw_setsockopt,
|
|
|
|
.getsockopt = raw_getsockopt,
|
|
|
|
.sendmsg = raw_sendmsg,
|
|
|
|
.recvmsg = raw_recvmsg,
|
|
|
|
.bind = raw_bind,
|
|
|
|
.backlog_rcv = raw_rcv_skb,
|
2013-01-21 09:00:03 +07:00
|
|
|
.release_cb = ip4_datagram_release_cb,
|
2008-03-23 06:56:51 +07:00
|
|
|
.hash = raw_hash_sk,
|
|
|
|
.unhash = raw_unhash_sk,
|
2006-03-21 13:48:35 +07:00
|
|
|
.obj_size = sizeof(struct raw_sock),
|
2008-03-23 06:56:51 +07:00
|
|
|
.h.raw_hash = &raw_v4_hashinfo,
|
2006-03-21 13:45:21 +07:00
|
|
|
#ifdef CONFIG_COMPAT
|
2006-03-21 13:48:35 +07:00
|
|
|
.compat_setsockopt = compat_raw_setsockopt,
|
|
|
|
.compat_getsockopt = compat_raw_getsockopt,
|
2011-01-29 23:15:56 +07:00
|
|
|
.compat_ioctl = compat_raw_ioctl,
|
2006-03-21 13:45:21 +07:00
|
|
|
#endif
|
2005-04-17 05:20:36 +07:00
|
|
|
};
|
|
|
|
|
|
|
|
#ifdef CONFIG_PROC_FS
|
|
|
|
static struct sock *raw_get_first(struct seq_file *seq)
|
|
|
|
{
|
|
|
|
struct sock *sk;
|
2008-11-03 15:24:34 +07:00
|
|
|
struct raw_iter_state *state = raw_seq_private(seq);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2007-11-20 13:36:45 +07:00
|
|
|
for (state->bucket = 0; state->bucket < RAW_HTABLE_SIZE;
|
|
|
|
++state->bucket) {
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:06:00 +07:00
|
|
|
sk_for_each(sk, &state->h->ht[state->bucket])
|
2008-03-26 00:36:06 +07:00
|
|
|
if (sock_net(sk) == seq_file_net(seq))
|
2005-04-17 05:20:36 +07:00
|
|
|
goto found;
|
|
|
|
}
|
|
|
|
sk = NULL;
|
|
|
|
found:
|
|
|
|
return sk;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct sock *raw_get_next(struct seq_file *seq, struct sock *sk)
|
|
|
|
{
|
2008-11-03 15:24:34 +07:00
|
|
|
struct raw_iter_state *state = raw_seq_private(seq);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
do {
|
|
|
|
sk = sk_next(sk);
|
|
|
|
try_again:
|
|
|
|
;
|
2008-03-26 00:36:06 +07:00
|
|
|
} while (sk && sock_net(sk) != seq_file_net(seq));
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2007-11-20 13:36:45 +07:00
|
|
|
if (!sk && ++state->bucket < RAW_HTABLE_SIZE) {
|
2007-11-20 13:38:33 +07:00
|
|
|
sk = sk_head(&state->h->ht[state->bucket]);
|
2005-04-17 05:20:36 +07:00
|
|
|
goto try_again;
|
|
|
|
}
|
|
|
|
return sk;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct sock *raw_get_idx(struct seq_file *seq, loff_t pos)
|
|
|
|
{
|
|
|
|
struct sock *sk = raw_get_first(seq);
|
|
|
|
|
|
|
|
if (sk)
|
|
|
|
while (pos && (sk = raw_get_next(seq, sk)) != NULL)
|
|
|
|
--pos;
|
|
|
|
return pos ? NULL : sk;
|
|
|
|
}
|
|
|
|
|
2007-11-20 13:38:33 +07:00
|
|
|
void *raw_seq_start(struct seq_file *seq, loff_t *pos)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2007-11-20 13:38:33 +07:00
|
|
|
struct raw_iter_state *state = raw_seq_private(seq);
|
|
|
|
|
|
|
|
read_lock(&state->h->lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
return *pos ? raw_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
|
|
|
|
}
|
2007-11-20 13:38:33 +07:00
|
|
|
EXPORT_SYMBOL_GPL(raw_seq_start);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2007-11-20 13:38:33 +07:00
|
|
|
void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
struct sock *sk;
|
|
|
|
|
|
|
|
if (v == SEQ_START_TOKEN)
|
|
|
|
sk = raw_get_first(seq);
|
|
|
|
else
|
|
|
|
sk = raw_get_next(seq, v);
|
|
|
|
++*pos;
|
|
|
|
return sk;
|
|
|
|
}
|
2007-11-20 13:38:33 +07:00
|
|
|
EXPORT_SYMBOL_GPL(raw_seq_next);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2007-11-20 13:38:33 +07:00
|
|
|
void raw_seq_stop(struct seq_file *seq, void *v)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2007-11-20 13:38:33 +07:00
|
|
|
struct raw_iter_state *state = raw_seq_private(seq);
|
|
|
|
|
|
|
|
read_unlock(&state->h->lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2007-11-20 13:38:33 +07:00
|
|
|
EXPORT_SYMBOL_GPL(raw_seq_stop);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-01-31 18:46:43 +07:00
|
|
|
static void raw_sock_seq_show(struct seq_file *seq, struct sock *sp, int i)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
struct inet_sock *inet = inet_sk(sp);
|
2009-10-15 13:30:45 +07:00
|
|
|
__be32 dest = inet->inet_daddr,
|
|
|
|
src = inet->inet_rcv_saddr;
|
2005-04-17 05:20:36 +07:00
|
|
|
__u16 destp = 0,
|
2009-10-15 13:30:45 +07:00
|
|
|
srcp = inet->inet_num;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-01-31 18:46:43 +07:00
|
|
|
seq_printf(seq, "%4d: %08X:%04X %08X:%04X"
|
2013-08-15 18:42:14 +07:00
|
|
|
" %02X %08X:%08X %02X:%08lX %08X %5u %8d %lu %d %pK %d\n",
|
2007-02-09 21:24:47 +07:00
|
|
|
i, src, srcp, dest, destp, sp->sk_state,
|
2009-06-18 09:05:41 +07:00
|
|
|
sk_wmem_alloc_get(sp),
|
|
|
|
sk_rmem_alloc_get(sp),
|
2012-05-24 14:10:10 +07:00
|
|
|
0, 0L, 0,
|
|
|
|
from_kuid_munged(seq_user_ns(seq), sock_i_uid(sp)),
|
|
|
|
0, sock_i_ino(sp),
|
2007-11-14 11:30:01 +07:00
|
|
|
atomic_read(&sp->sk_refcnt), sp, atomic_read(&sp->sk_drops));
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
static int raw_seq_show(struct seq_file *seq, void *v)
|
|
|
|
{
|
|
|
|
if (v == SEQ_START_TOKEN)
|
2008-01-31 18:46:43 +07:00
|
|
|
seq_printf(seq, " sl local_address rem_address st tx_queue "
|
|
|
|
"rx_queue tr tm->when retrnsmt uid timeout "
|
2008-06-18 11:04:56 +07:00
|
|
|
"inode ref pointer drops\n");
|
2008-01-31 18:46:43 +07:00
|
|
|
else
|
|
|
|
raw_sock_seq_show(seq, v, raw_seq_private(seq)->bucket);
|
2005-04-17 05:20:36 +07:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-03-13 04:34:29 +07:00
|
|
|
static const struct seq_operations raw_seq_ops = {
|
2005-04-17 05:20:36 +07:00
|
|
|
.start = raw_seq_start,
|
|
|
|
.next = raw_seq_next,
|
|
|
|
.stop = raw_seq_stop,
|
|
|
|
.show = raw_seq_show,
|
|
|
|
};
|
|
|
|
|
2008-01-31 18:48:55 +07:00
|
|
|
int raw_seq_open(struct inode *ino, struct file *file,
|
|
|
|
struct raw_hashinfo *h, const struct seq_operations *ops)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-01-14 20:35:57 +07:00
|
|
|
int err;
|
2007-11-20 13:38:33 +07:00
|
|
|
struct raw_iter_state *i;
|
|
|
|
|
2008-01-31 18:48:55 +07:00
|
|
|
err = seq_open_net(ino, file, ops, sizeof(struct raw_iter_state));
|
2008-01-14 20:35:57 +07:00
|
|
|
if (err < 0)
|
|
|
|
return err;
|
2007-11-20 13:38:33 +07:00
|
|
|
|
2008-01-14 20:35:57 +07:00
|
|
|
i = raw_seq_private((struct seq_file *)file->private_data);
|
2007-11-20 13:38:33 +07:00
|
|
|
i->h = h;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(raw_seq_open);
|
|
|
|
|
|
|
|
static int raw_v4_seq_open(struct inode *inode, struct file *file)
|
|
|
|
{
|
2008-01-31 18:48:55 +07:00
|
|
|
return raw_seq_open(inode, file, &raw_v4_hashinfo, &raw_seq_ops);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2007-02-12 15:55:35 +07:00
|
|
|
static const struct file_operations raw_seq_fops = {
|
2005-04-17 05:20:36 +07:00
|
|
|
.owner = THIS_MODULE,
|
2007-11-20 13:38:33 +07:00
|
|
|
.open = raw_v4_seq_open,
|
2005-04-17 05:20:36 +07:00
|
|
|
.read = seq_read,
|
|
|
|
.llseek = seq_lseek,
|
2008-01-14 20:35:57 +07:00
|
|
|
.release = seq_release_net,
|
2005-04-17 05:20:36 +07:00
|
|
|
};
|
|
|
|
|
2008-01-14 20:36:50 +07:00
|
|
|
static __net_init int raw_init_net(struct net *net)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2013-02-18 08:34:54 +07:00
|
|
|
if (!proc_create("raw", S_IRUGO, net->proc_net, &raw_seq_fops))
|
2005-04-17 05:20:36 +07:00
|
|
|
return -ENOMEM;
|
2008-01-14 20:36:50 +07:00
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-01-14 20:36:50 +07:00
|
|
|
static __net_exit void raw_exit_net(struct net *net)
|
|
|
|
{
|
2013-02-18 08:34:56 +07:00
|
|
|
remove_proc_entry("raw", net->proc_net);
|
2008-01-14 20:36:50 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
static __net_initdata struct pernet_operations raw_net_ops = {
|
|
|
|
.init = raw_init_net,
|
|
|
|
.exit = raw_exit_net,
|
|
|
|
};
|
|
|
|
|
|
|
|
int __init raw_proc_init(void)
|
|
|
|
{
|
|
|
|
return register_pernet_subsys(&raw_net_ops);
|
|
|
|
}
|
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
void __init raw_proc_exit(void)
|
|
|
|
{
|
2008-01-14 20:36:50 +07:00
|
|
|
unregister_pernet_subsys(&raw_net_ops);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
#endif /* CONFIG_PROC_FS */
|