2019-05-27 14:55:01 +08:00
|
|
|
/* SPDX-License-Identifier: GPL-2.0-or-later */
|
2005-04-17 06:20:36 +08:00
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*
|
|
|
|
* Definitions for the RAW-IP module.
|
|
|
|
*
|
|
|
|
* Version: @(#)raw.h 1.0.2 05/07/93
|
|
|
|
*
|
|
|
|
* Author: Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
|
|
|
|
*/
|
|
|
|
#ifndef _RAW_H
|
|
|
|
#define _RAW_H
|
|
|
|
|
2018-11-07 23:36:06 +08:00
|
|
|
#include <net/inet_sock.h>
|
2005-12-27 12:43:12 +08:00
|
|
|
#include <net/protocol.h>
|
2023-02-02 17:41:00 +08:00
|
|
|
#include <net/netns/hash.h>
|
|
|
|
#include <linux/hash.h>
|
2010-04-13 13:03:18 +08:00
|
|
|
#include <linux/icmp.h>
|
2005-12-27 12:43:12 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
extern struct proto raw_prot;
|
|
|
|
|
2016-10-21 18:03:44 +08:00
|
|
|
extern struct raw_hashinfo raw_v4_hashinfo;
|
2023-03-16 23:32:01 +08:00
|
|
|
bool raw_v4_match(struct net *net, const struct sock *sk, unsigned short num,
|
2022-06-18 11:47:04 +08:00
|
|
|
__be32 raddr, __be32 laddr, int dif, int sdif);
|
2016-10-21 18:03:44 +08:00
|
|
|
|
|
|
|
int raw_abort(struct sock *sk, int err);
|
2007-11-20 14:35:07 +08:00
|
|
|
void raw_icmp_error(struct sk_buff *, int, u32);
|
|
|
|
int raw_local_deliver(struct sk_buff *, int);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2013-09-23 01:32:19 +08:00
|
|
|
int raw_rcv(struct sock *, struct sk_buff *);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2023-02-02 17:41:00 +08:00
|
|
|
#define RAW_HTABLE_LOG 8
|
|
|
|
#define RAW_HTABLE_SIZE (1U << RAW_HTABLE_LOG)
|
2007-11-20 14:36:45 +08:00
|
|
|
|
|
|
|
struct raw_hashinfo {
|
2022-06-20 18:05:09 +08:00
|
|
|
spinlock_t lock;
|
2023-02-02 17:41:00 +08:00
|
|
|
|
raw: Fix NULL deref in raw_get_next().
Dae R. Jeong reported a NULL deref in raw_get_next() [0].
It seems that the repro was running these sequences in parallel so
that one thread was iterating on a socket that was being freed in
another netns.
unshare(0x40060200)
r0 = syz_open_procfs(0x0, &(0x7f0000002080)='net/raw\x00')
socket$inet_icmp_raw(0x2, 0x3, 0x1)
pread64(r0, &(0x7f0000000000)=""/10, 0xa, 0x10000000007f)
After commit 0daf07e52709 ("raw: convert raw sockets to RCU"), we
use RCU and hlist_nulls_for_each_entry() to iterate over SOCK_RAW
sockets. However, we should use spinlock for slow paths to avoid
the NULL deref.
Also, SOCK_RAW does not use SLAB_TYPESAFE_BY_RCU, and the slab object
is not reused during iteration in the grace period. In fact, the
lockless readers do not check the nulls marker with get_nulls_value().
So, SOCK_RAW should use hlist instead of hlist_nulls.
Instead of adding an unnecessary barrier by sk_nulls_for_each_rcu(),
let's convert hlist_nulls to hlist and use sk_for_each_rcu() for
fast paths and sk_for_each() and spinlock for /proc/net/raw.
[0]:
general protection fault, probably for non-canonical address 0xdffffc0000000005: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f]
CPU: 2 PID: 20952 Comm: syz-executor.0 Not tainted 6.2.0-g048ec869bafd-dirty #7
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014
RIP: 0010:read_pnet include/net/net_namespace.h:383 [inline]
RIP: 0010:sock_net include/net/sock.h:649 [inline]
RIP: 0010:raw_get_next net/ipv4/raw.c:974 [inline]
RIP: 0010:raw_get_idx net/ipv4/raw.c:986 [inline]
RIP: 0010:raw_seq_start+0x431/0x800 net/ipv4/raw.c:995
Code: ef e8 33 3d 94 f7 49 8b 6d 00 4c 89 ef e8 b7 65 5f f7 49 89 ed 49 83 c5 98 0f 84 9a 00 00 00 48 83 c5 c8 48 89 e8 48 c1 e8 03 <42> 80 3c 30 00 74 08 48 89 ef e8 00 3d 94 f7 4c 8b 7d 00 48 89 ef
RSP: 0018:ffffc9001154f9b0 EFLAGS: 00010206
RAX: 0000000000000005 RBX: 1ffff1100302c8fd RCX: 0000000000000000
RDX: 0000000000000028 RSI: ffffc9001154f988 RDI: ffffc9000f77a338
RBP: 0000000000000029 R08: ffffffff8a50ffb4 R09: fffffbfff24b6bd9
R10: fffffbfff24b6bd9 R11: 0000000000000000 R12: ffff88801db73b78
R13: fffffffffffffff9 R14: dffffc0000000000 R15: 0000000000000030
FS: 00007f843ae8e700(0000) GS:ffff888063700000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000055bb9614b35f CR3: 000000003c672000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
seq_read_iter+0x4c6/0x10f0 fs/seq_file.c:225
seq_read+0x224/0x320 fs/seq_file.c:162
pde_read fs/proc/inode.c:316 [inline]
proc_reg_read+0x23f/0x330 fs/proc/inode.c:328
vfs_read+0x31e/0xd30 fs/read_write.c:468
ksys_pread64 fs/read_write.c:665 [inline]
__do_sys_pread64 fs/read_write.c:675 [inline]
__se_sys_pread64 fs/read_write.c:672 [inline]
__x64_sys_pread64+0x1e9/0x280 fs/read_write.c:672
do_syscall_x64 arch/x86/entry/common.c:51 [inline]
do_syscall_64+0x4e/0xa0 arch/x86/entry/common.c:82
entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x478d29
Code: f7 d8 64 89 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f843ae8dbe8 EFLAGS: 00000246 ORIG_RAX: 0000000000000011
RAX: ffffffffffffffda RBX: 0000000000791408 RCX: 0000000000478d29
RDX: 000000000000000a RSI: 0000000020000000 RDI: 0000000000000003
RBP: 00000000f477909a R08: 0000000000000000 R09: 0000000000000000
R10: 000010000000007f R11: 0000000000000246 R12: 0000000000791740
R13: 0000000000791414 R14: 0000000000791408 R15: 00007ffc2eb48a50
</TASK>
Modules linked in:
---[ end trace 0000000000000000 ]---
RIP: 0010:read_pnet include/net/net_namespace.h:383 [inline]
RIP: 0010:sock_net include/net/sock.h:649 [inline]
RIP: 0010:raw_get_next net/ipv4/raw.c:974 [inline]
RIP: 0010:raw_get_idx net/ipv4/raw.c:986 [inline]
RIP: 0010:raw_seq_start+0x431/0x800 net/ipv4/raw.c:995
Code: ef e8 33 3d 94 f7 49 8b 6d 00 4c 89 ef e8 b7 65 5f f7 49 89 ed 49 83 c5 98 0f 84 9a 00 00 00 48 83 c5 c8 48 89 e8 48 c1 e8 03 <42> 80 3c 30 00 74 08 48 89 ef e8 00 3d 94 f7 4c 8b 7d 00 48 89 ef
RSP: 0018:ffffc9001154f9b0 EFLAGS: 00010206
RAX: 0000000000000005 RBX: 1ffff1100302c8fd RCX: 0000000000000000
RDX: 0000000000000028 RSI: ffffc9001154f988 RDI: ffffc9000f77a338
RBP: 0000000000000029 R08: ffffffff8a50ffb4 R09: fffffbfff24b6bd9
R10: fffffbfff24b6bd9 R11: 0000000000000000 R12: ffff88801db73b78
R13: fffffffffffffff9 R14: dffffc0000000000 R15: 0000000000000030
FS: 00007f843ae8e700(0000) GS:ffff888063700000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f92ff166000 CR3: 000000003c672000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Fixes: 0daf07e52709 ("raw: convert raw sockets to RCU")
Reported-by: syzbot <syzkaller@googlegroups.com>
Reported-by: Dae R. Jeong <threeearcat@gmail.com>
Link: https://lore.kernel.org/netdev/ZCA2mGV_cmq7lIfV@dragonet/
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2023-04-04 03:49:58 +08:00
|
|
|
struct hlist_head ht[RAW_HTABLE_SIZE] ____cacheline_aligned;
|
2007-11-20 14:36:45 +08:00
|
|
|
};
|
|
|
|
|
2023-02-02 17:41:00 +08:00
|
|
|
static inline u32 raw_hashfunc(const struct net *net, u32 proto)
|
|
|
|
{
|
|
|
|
return hash_32(net_hash_mix(net) ^ proto, RAW_HTABLE_LOG);
|
|
|
|
}
|
|
|
|
|
2022-06-18 11:47:05 +08:00
|
|
|
static inline void raw_hashinfo_init(struct raw_hashinfo *hashinfo)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
2022-06-20 18:05:09 +08:00
|
|
|
spin_lock_init(&hashinfo->lock);
|
2022-06-18 11:47:05 +08:00
|
|
|
for (i = 0; i < RAW_HTABLE_SIZE; i++)
|
raw: Fix NULL deref in raw_get_next().
Dae R. Jeong reported a NULL deref in raw_get_next() [0].
It seems that the repro was running these sequences in parallel so
that one thread was iterating on a socket that was being freed in
another netns.
unshare(0x40060200)
r0 = syz_open_procfs(0x0, &(0x7f0000002080)='net/raw\x00')
socket$inet_icmp_raw(0x2, 0x3, 0x1)
pread64(r0, &(0x7f0000000000)=""/10, 0xa, 0x10000000007f)
After commit 0daf07e52709 ("raw: convert raw sockets to RCU"), we
use RCU and hlist_nulls_for_each_entry() to iterate over SOCK_RAW
sockets. However, we should use spinlock for slow paths to avoid
the NULL deref.
Also, SOCK_RAW does not use SLAB_TYPESAFE_BY_RCU, and the slab object
is not reused during iteration in the grace period. In fact, the
lockless readers do not check the nulls marker with get_nulls_value().
So, SOCK_RAW should use hlist instead of hlist_nulls.
Instead of adding an unnecessary barrier by sk_nulls_for_each_rcu(),
let's convert hlist_nulls to hlist and use sk_for_each_rcu() for
fast paths and sk_for_each() and spinlock for /proc/net/raw.
[0]:
general protection fault, probably for non-canonical address 0xdffffc0000000005: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f]
CPU: 2 PID: 20952 Comm: syz-executor.0 Not tainted 6.2.0-g048ec869bafd-dirty #7
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014
RIP: 0010:read_pnet include/net/net_namespace.h:383 [inline]
RIP: 0010:sock_net include/net/sock.h:649 [inline]
RIP: 0010:raw_get_next net/ipv4/raw.c:974 [inline]
RIP: 0010:raw_get_idx net/ipv4/raw.c:986 [inline]
RIP: 0010:raw_seq_start+0x431/0x800 net/ipv4/raw.c:995
Code: ef e8 33 3d 94 f7 49 8b 6d 00 4c 89 ef e8 b7 65 5f f7 49 89 ed 49 83 c5 98 0f 84 9a 00 00 00 48 83 c5 c8 48 89 e8 48 c1 e8 03 <42> 80 3c 30 00 74 08 48 89 ef e8 00 3d 94 f7 4c 8b 7d 00 48 89 ef
RSP: 0018:ffffc9001154f9b0 EFLAGS: 00010206
RAX: 0000000000000005 RBX: 1ffff1100302c8fd RCX: 0000000000000000
RDX: 0000000000000028 RSI: ffffc9001154f988 RDI: ffffc9000f77a338
RBP: 0000000000000029 R08: ffffffff8a50ffb4 R09: fffffbfff24b6bd9
R10: fffffbfff24b6bd9 R11: 0000000000000000 R12: ffff88801db73b78
R13: fffffffffffffff9 R14: dffffc0000000000 R15: 0000000000000030
FS: 00007f843ae8e700(0000) GS:ffff888063700000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000055bb9614b35f CR3: 000000003c672000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
seq_read_iter+0x4c6/0x10f0 fs/seq_file.c:225
seq_read+0x224/0x320 fs/seq_file.c:162
pde_read fs/proc/inode.c:316 [inline]
proc_reg_read+0x23f/0x330 fs/proc/inode.c:328
vfs_read+0x31e/0xd30 fs/read_write.c:468
ksys_pread64 fs/read_write.c:665 [inline]
__do_sys_pread64 fs/read_write.c:675 [inline]
__se_sys_pread64 fs/read_write.c:672 [inline]
__x64_sys_pread64+0x1e9/0x280 fs/read_write.c:672
do_syscall_x64 arch/x86/entry/common.c:51 [inline]
do_syscall_64+0x4e/0xa0 arch/x86/entry/common.c:82
entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x478d29
Code: f7 d8 64 89 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f843ae8dbe8 EFLAGS: 00000246 ORIG_RAX: 0000000000000011
RAX: ffffffffffffffda RBX: 0000000000791408 RCX: 0000000000478d29
RDX: 000000000000000a RSI: 0000000020000000 RDI: 0000000000000003
RBP: 00000000f477909a R08: 0000000000000000 R09: 0000000000000000
R10: 000010000000007f R11: 0000000000000246 R12: 0000000000791740
R13: 0000000000791414 R14: 0000000000791408 R15: 00007ffc2eb48a50
</TASK>
Modules linked in:
---[ end trace 0000000000000000 ]---
RIP: 0010:read_pnet include/net/net_namespace.h:383 [inline]
RIP: 0010:sock_net include/net/sock.h:649 [inline]
RIP: 0010:raw_get_next net/ipv4/raw.c:974 [inline]
RIP: 0010:raw_get_idx net/ipv4/raw.c:986 [inline]
RIP: 0010:raw_seq_start+0x431/0x800 net/ipv4/raw.c:995
Code: ef e8 33 3d 94 f7 49 8b 6d 00 4c 89 ef e8 b7 65 5f f7 49 89 ed 49 83 c5 98 0f 84 9a 00 00 00 48 83 c5 c8 48 89 e8 48 c1 e8 03 <42> 80 3c 30 00 74 08 48 89 ef e8 00 3d 94 f7 4c 8b 7d 00 48 89 ef
RSP: 0018:ffffc9001154f9b0 EFLAGS: 00010206
RAX: 0000000000000005 RBX: 1ffff1100302c8fd RCX: 0000000000000000
RDX: 0000000000000028 RSI: ffffc9001154f988 RDI: ffffc9000f77a338
RBP: 0000000000000029 R08: ffffffff8a50ffb4 R09: fffffbfff24b6bd9
R10: fffffbfff24b6bd9 R11: 0000000000000000 R12: ffff88801db73b78
R13: fffffffffffffff9 R14: dffffc0000000000 R15: 0000000000000030
FS: 00007f843ae8e700(0000) GS:ffff888063700000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f92ff166000 CR3: 000000003c672000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Fixes: 0daf07e52709 ("raw: convert raw sockets to RCU")
Reported-by: syzbot <syzkaller@googlegroups.com>
Reported-by: Dae R. Jeong <threeearcat@gmail.com>
Link: https://lore.kernel.org/netdev/ZCA2mGV_cmq7lIfV@dragonet/
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2023-04-04 03:49:58 +08:00
|
|
|
INIT_HLIST_HEAD(&hashinfo->ht[i]);
|
2022-06-18 11:47:05 +08:00
|
|
|
}
|
|
|
|
|
2005-08-16 13:18:02 +08:00
|
|
|
#ifdef CONFIG_PROC_FS
|
2013-09-23 01:32:19 +08:00
|
|
|
int raw_proc_init(void);
|
|
|
|
void raw_proc_exit(void);
|
2007-11-20 14:38:33 +08:00
|
|
|
|
|
|
|
struct raw_iter_state {
|
2008-01-14 21:35:57 +08:00
|
|
|
struct seq_net_private p;
|
2007-11-20 14:38:33 +08:00
|
|
|
int bucket;
|
|
|
|
};
|
|
|
|
|
2010-09-07 13:55:38 +08:00
|
|
|
static inline struct raw_iter_state *raw_seq_private(struct seq_file *seq)
|
|
|
|
{
|
|
|
|
return seq->private;
|
|
|
|
}
|
2007-11-20 14:38:33 +08:00
|
|
|
void *raw_seq_start(struct seq_file *seq, loff_t *pos);
|
|
|
|
void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos);
|
|
|
|
void raw_seq_stop(struct seq_file *seq, void *v);
|
2005-08-16 13:18:02 +08:00
|
|
|
#endif
|
|
|
|
|
2016-02-11 00:50:35 +08:00
|
|
|
int raw_hash_sk(struct sock *sk);
|
2008-03-23 07:56:51 +08:00
|
|
|
void raw_unhash_sk(struct sock *sk);
|
2018-11-07 23:36:05 +08:00
|
|
|
void raw_init(void);
|
2007-11-20 14:37:24 +08:00
|
|
|
|
2010-04-13 13:03:18 +08:00
|
|
|
struct raw_sock {
|
|
|
|
/* inet_sock has to be the first member */
|
|
|
|
struct inet_sock inet;
|
|
|
|
struct icmp_filter filter;
|
ipv4: ipmr: support multiple tables
This patch adds support for multiple independant multicast routing instances,
named "tables".
Userspace multicast routing daemons can bind to a specific table instance by
issuing a setsockopt call using a new option MRT_TABLE. The table number is
stored in the raw socket data and affects all following ipmr setsockopt(),
getsockopt() and ioctl() calls. By default, a single table (RT_TABLE_DEFAULT)
is created with a default routing rule pointing to it. Newly created pimreg
devices have the table number appended ("pimregX"), with the exception of
devices created in the default table, which are named just "pimreg" for
compatibility reasons.
Packets are directed to a specific table instance using routing rules,
similar to how regular routing rules work. Currently iif, oif and mark
are supported as keys, source and destination addresses could be supported
additionally.
Example usage:
- bind pimd/xorp/... to a specific table:
uint32_t table = 123;
setsockopt(fd, IPPROTO_IP, MRT_TABLE, &table, sizeof(table));
- create routing rules directing packets to the new table:
# ip mrule add iif eth0 lookup 123
# ip mrule add oif eth0 lookup 123
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-13 13:03:23 +08:00
|
|
|
u32 ipmr_table;
|
2010-04-13 13:03:18 +08:00
|
|
|
};
|
|
|
|
|
2023-03-17 23:55:32 +08:00
|
|
|
#define raw_sk(ptr) container_of_const(ptr, struct raw_sock, inet.sk)
|
2010-04-13 13:03:18 +08:00
|
|
|
|
2018-11-07 23:36:06 +08:00
|
|
|
static inline bool raw_sk_bound_dev_eq(struct net *net, int bound_dev_if,
|
|
|
|
int dif, int sdif)
|
|
|
|
{
|
|
|
|
#if IS_ENABLED(CONFIG_NET_L3_MASTER_DEV)
|
2022-07-12 08:15:29 +08:00
|
|
|
return inet_bound_dev_eq(READ_ONCE(net->ipv4.sysctl_raw_l3mdev_accept),
|
2018-11-07 23:36:06 +08:00
|
|
|
bound_dev_if, dif, sdif);
|
|
|
|
#else
|
|
|
|
return inet_bound_dev_eq(true, bound_dev_if, dif, sdif);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
#endif /* _RAW_H */
|