linux/net/rds/bind.c

284 lines
7.7 KiB
C
Raw Normal View History

/*
* Copyright (c) 2006, 2019 Oracle and/or its affiliates. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*
*/
#include <linux/kernel.h>
#include <net/sock.h>
#include <linux/in.h>
#include <linux/ipv6.h>
#include <linux/if_arp.h>
#include <linux/jhash.h>
#include <linux/ratelimit.h>
#include "rds.h"
static struct rhashtable bind_hash_table;
static const struct rhashtable_params ht_parms = {
.nelem_hint = 768,
.key_len = RDS_BOUND_KEY_LEN,
.key_offset = offsetof(struct rds_sock, rs_bound_key),
.head_offset = offsetof(struct rds_sock, rs_bound_node),
.max_size = 16384,
.min_size = 1024,
};
/* Create a key for the bind hash table manipulation. Port is in network byte
* order.
*/
static inline void __rds_create_bind_key(u8 *key, const struct in6_addr *addr,
__be16 port, __u32 scope_id)
{
memcpy(key, addr, sizeof(*addr));
key += sizeof(*addr);
memcpy(key, &port, sizeof(port));
key += sizeof(port);
memcpy(key, &scope_id, sizeof(scope_id));
}
/*
* Return the rds_sock bound at the given local address.
*
* The rx path can race with rds_release. We notice if rds_release() has
* marked this socket and don't return a rs ref to the rx path.
*/
struct rds_sock *rds_find_bound(const struct in6_addr *addr, __be16 port,
__u32 scope_id)
{
u8 key[RDS_BOUND_KEY_LEN];
struct rds_sock *rs;
__rds_create_bind_key(key, addr, port, scope_id);
rcu_read_lock();
rs = rhashtable_lookup(&bind_hash_table, key, ht_parms);
rds: fix refcount bug in rds_sock_addref syzbot was able to catch a bug in rds [1] The issue here is that the socket might be found in a hash table but that its refcount has already be set to 0 by another cpu. We need to use refcount_inc_not_zero() to be safe here. [1] refcount_t: increment on 0; use-after-free. WARNING: CPU: 1 PID: 23129 at lib/refcount.c:153 refcount_inc_checked lib/refcount.c:153 [inline] WARNING: CPU: 1 PID: 23129 at lib/refcount.c:153 refcount_inc_checked+0x61/0x70 lib/refcount.c:151 Kernel panic - not syncing: panic_on_warn set ... CPU: 1 PID: 23129 Comm: syz-executor3 Not tainted 5.0.0-rc4+ #53 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x1db/0x2d0 lib/dump_stack.c:113 panic+0x2cb/0x65c kernel/panic.c:214 __warn.cold+0x20/0x48 kernel/panic.c:571 report_bug+0x263/0x2b0 lib/bug.c:186 fixup_bug arch/x86/kernel/traps.c:178 [inline] fixup_bug arch/x86/kernel/traps.c:173 [inline] do_error_trap+0x11b/0x200 arch/x86/kernel/traps.c:271 do_invalid_op+0x37/0x50 arch/x86/kernel/traps.c:290 invalid_op+0x14/0x20 arch/x86/entry/entry_64.S:973 RIP: 0010:refcount_inc_checked lib/refcount.c:153 [inline] RIP: 0010:refcount_inc_checked+0x61/0x70 lib/refcount.c:151 Code: 1d 51 63 c8 06 31 ff 89 de e8 eb 1b f2 fd 84 db 75 dd e8 a2 1a f2 fd 48 c7 c7 60 9f 81 88 c6 05 31 63 c8 06 01 e8 af 65 bb fd <0f> 0b eb c1 90 66 2e 0f 1f 84 00 00 00 00 00 55 48 89 e5 41 54 49 RSP: 0018:ffff8880a0cbf1e8 EFLAGS: 00010282 RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffc90006113000 RDX: 000000000001047d RSI: ffffffff81685776 RDI: 0000000000000005 RBP: ffff8880a0cbf1f8 R08: ffff888097c9e100 R09: ffffed1015ce5021 R10: ffffed1015ce5020 R11: ffff8880ae728107 R12: ffff8880723c20c0 R13: ffff8880723c24b0 R14: dffffc0000000000 R15: ffffed1014197e64 sock_hold include/net/sock.h:647 [inline] rds_sock_addref+0x19/0x20 net/rds/af_rds.c:675 rds_find_bound+0x97c/0x1080 net/rds/bind.c:82 rds_recv_incoming+0x3be/0x1430 net/rds/recv.c:362 rds_loop_xmit+0xf3/0x2a0 net/rds/loop.c:96 rds_send_xmit+0x1355/0x2a10 net/rds/send.c:355 rds_sendmsg+0x323c/0x44e0 net/rds/send.c:1368 sock_sendmsg_nosec net/socket.c:621 [inline] sock_sendmsg+0xdd/0x130 net/socket.c:631 __sys_sendto+0x387/0x5f0 net/socket.c:1788 __do_sys_sendto net/socket.c:1800 [inline] __se_sys_sendto net/socket.c:1796 [inline] __x64_sys_sendto+0xe1/0x1a0 net/socket.c:1796 do_syscall_64+0x1a3/0x800 arch/x86/entry/common.c:290 entry_SYSCALL_64_after_hwframe+0x49/0xbe RIP: 0033:0x458089 Code: 6d b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 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 0f 83 3b b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 RSP: 002b:00007fc266df8c78 EFLAGS: 00000246 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 0000000000000006 RCX: 0000000000458089 RDX: 0000000000000000 RSI: 00000000204b3fff RDI: 0000000000000005 RBP: 000000000073bf00 R08: 00000000202b4000 R09: 0000000000000010 R10: 0000000000000000 R11: 0000000000000246 R12: 00007fc266df96d4 R13: 00000000004c56e4 R14: 00000000004d94a8 R15: 00000000ffffffff Fixes: cc4dfb7f70a3 ("rds: fix two RCU related problems") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Sowmini Varadhan <sowmini.varadhan@oracle.com> Cc: Santosh Shilimkar <santosh.shilimkar@oracle.com> Cc: rds-devel@oss.oracle.com Cc: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-31 16:47:10 +00:00
if (rs && (sock_flag(rds_rs_to_sk(rs), SOCK_DEAD) ||
!refcount_inc_not_zero(&rds_rs_to_sk(rs)->sk_refcnt)))
rs = NULL;
rds: fix refcount bug in rds_sock_addref syzbot was able to catch a bug in rds [1] The issue here is that the socket might be found in a hash table but that its refcount has already be set to 0 by another cpu. We need to use refcount_inc_not_zero() to be safe here. [1] refcount_t: increment on 0; use-after-free. WARNING: CPU: 1 PID: 23129 at lib/refcount.c:153 refcount_inc_checked lib/refcount.c:153 [inline] WARNING: CPU: 1 PID: 23129 at lib/refcount.c:153 refcount_inc_checked+0x61/0x70 lib/refcount.c:151 Kernel panic - not syncing: panic_on_warn set ... CPU: 1 PID: 23129 Comm: syz-executor3 Not tainted 5.0.0-rc4+ #53 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x1db/0x2d0 lib/dump_stack.c:113 panic+0x2cb/0x65c kernel/panic.c:214 __warn.cold+0x20/0x48 kernel/panic.c:571 report_bug+0x263/0x2b0 lib/bug.c:186 fixup_bug arch/x86/kernel/traps.c:178 [inline] fixup_bug arch/x86/kernel/traps.c:173 [inline] do_error_trap+0x11b/0x200 arch/x86/kernel/traps.c:271 do_invalid_op+0x37/0x50 arch/x86/kernel/traps.c:290 invalid_op+0x14/0x20 arch/x86/entry/entry_64.S:973 RIP: 0010:refcount_inc_checked lib/refcount.c:153 [inline] RIP: 0010:refcount_inc_checked+0x61/0x70 lib/refcount.c:151 Code: 1d 51 63 c8 06 31 ff 89 de e8 eb 1b f2 fd 84 db 75 dd e8 a2 1a f2 fd 48 c7 c7 60 9f 81 88 c6 05 31 63 c8 06 01 e8 af 65 bb fd <0f> 0b eb c1 90 66 2e 0f 1f 84 00 00 00 00 00 55 48 89 e5 41 54 49 RSP: 0018:ffff8880a0cbf1e8 EFLAGS: 00010282 RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffc90006113000 RDX: 000000000001047d RSI: ffffffff81685776 RDI: 0000000000000005 RBP: ffff8880a0cbf1f8 R08: ffff888097c9e100 R09: ffffed1015ce5021 R10: ffffed1015ce5020 R11: ffff8880ae728107 R12: ffff8880723c20c0 R13: ffff8880723c24b0 R14: dffffc0000000000 R15: ffffed1014197e64 sock_hold include/net/sock.h:647 [inline] rds_sock_addref+0x19/0x20 net/rds/af_rds.c:675 rds_find_bound+0x97c/0x1080 net/rds/bind.c:82 rds_recv_incoming+0x3be/0x1430 net/rds/recv.c:362 rds_loop_xmit+0xf3/0x2a0 net/rds/loop.c:96 rds_send_xmit+0x1355/0x2a10 net/rds/send.c:355 rds_sendmsg+0x323c/0x44e0 net/rds/send.c:1368 sock_sendmsg_nosec net/socket.c:621 [inline] sock_sendmsg+0xdd/0x130 net/socket.c:631 __sys_sendto+0x387/0x5f0 net/socket.c:1788 __do_sys_sendto net/socket.c:1800 [inline] __se_sys_sendto net/socket.c:1796 [inline] __x64_sys_sendto+0xe1/0x1a0 net/socket.c:1796 do_syscall_64+0x1a3/0x800 arch/x86/entry/common.c:290 entry_SYSCALL_64_after_hwframe+0x49/0xbe RIP: 0033:0x458089 Code: 6d b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 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 0f 83 3b b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 RSP: 002b:00007fc266df8c78 EFLAGS: 00000246 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 0000000000000006 RCX: 0000000000458089 RDX: 0000000000000000 RSI: 00000000204b3fff RDI: 0000000000000005 RBP: 000000000073bf00 R08: 00000000202b4000 R09: 0000000000000010 R10: 0000000000000000 R11: 0000000000000246 R12: 00007fc266df96d4 R13: 00000000004c56e4 R14: 00000000004d94a8 R15: 00000000ffffffff Fixes: cc4dfb7f70a3 ("rds: fix two RCU related problems") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot <syzkaller@googlegroups.com> Cc: Sowmini Varadhan <sowmini.varadhan@oracle.com> Cc: Santosh Shilimkar <santosh.shilimkar@oracle.com> Cc: rds-devel@oss.oracle.com Cc: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-31 16:47:10 +00:00
rcu_read_unlock();
rdsdebug("returning rs %p for %pI6c:%u\n", rs, addr,
ntohs(port));
return rs;
}
/* returns -ve errno or +ve port */
static int rds_add_bound(struct rds_sock *rs, const struct in6_addr *addr,
__be16 *port, __u32 scope_id)
{
int ret = -EADDRINUSE;
u16 rover, last;
u8 key[RDS_BOUND_KEY_LEN];
if (*port != 0) {
rover = be16_to_cpu(*port);
if (rover == RDS_FLAG_PROBE_PORT)
return -EINVAL;
last = rover;
} else {
rover = max_t(u16, prandom_u32(), 2);
last = rover - 1;
}
do {
if (rover == 0)
rover++;
if (rover == RDS_FLAG_PROBE_PORT)
continue;
__rds_create_bind_key(key, addr, cpu_to_be16(rover),
scope_id);
if (rhashtable_lookup_fast(&bind_hash_table, key, ht_parms))
continue;
memcpy(rs->rs_bound_key, key, sizeof(rs->rs_bound_key));
rs->rs_bound_addr = *addr;
net_get_random_once(&rs->rs_hash_initval,
sizeof(rs->rs_hash_initval));
rs->rs_bound_port = cpu_to_be16(rover);
rs->rs_bound_node.next = NULL;
rds_sock_addref(rs);
if (!rhashtable_insert_fast(&bind_hash_table,
&rs->rs_bound_node, ht_parms)) {
*port = rs->rs_bound_port;
rs->rs_bound_scope_id = scope_id;
ret = 0;
rdsdebug("rs %p binding to %pI6c:%d\n",
rs, addr, (int)ntohs(*port));
break;
} else {
rs->rs_bound_addr = in6addr_any;
rds_sock_put(rs);
ret = -ENOMEM;
break;
}
} while (rover++ != last);
return ret;
}
void rds_remove_bound(struct rds_sock *rs)
{
if (ipv6_addr_any(&rs->rs_bound_addr))
return;
rdsdebug("rs %p unbinding from %pI6c:%d\n",
rs, &rs->rs_bound_addr,
ntohs(rs->rs_bound_port));
rhashtable_remove_fast(&bind_hash_table, &rs->rs_bound_node, ht_parms);
rds_sock_put(rs);
rs->rs_bound_addr = in6addr_any;
}
int rds_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
{
struct sock *sk = sock->sk;
struct rds_sock *rs = rds_sk_to_rs(sk);
struct in6_addr v6addr, *binding_addr;
struct rds_transport *trans;
__u32 scope_id = 0;
int ret = 0;
__be16 port;
/* We allow an RDS socket to be bound to either IPv4 or IPv6
* address.
*/
if (addr_len < offsetofend(struct sockaddr, sa_family))
return -EINVAL;
if (uaddr->sa_family == AF_INET) {
struct sockaddr_in *sin = (struct sockaddr_in *)uaddr;
if (addr_len < sizeof(struct sockaddr_in) ||
sin->sin_addr.s_addr == htonl(INADDR_ANY) ||
sin->sin_addr.s_addr == htonl(INADDR_BROADCAST) ||
ipv4_is_multicast(sin->sin_addr.s_addr))
return -EINVAL;
ipv6_addr_set_v4mapped(sin->sin_addr.s_addr, &v6addr);
binding_addr = &v6addr;
port = sin->sin_port;
#if IS_ENABLED(CONFIG_IPV6)
} else if (uaddr->sa_family == AF_INET6) {
struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)uaddr;
int addr_type;
if (addr_len < sizeof(struct sockaddr_in6))
return -EINVAL;
addr_type = ipv6_addr_type(&sin6->sin6_addr);
if (!(addr_type & IPV6_ADDR_UNICAST)) {
__be32 addr4;
if (!(addr_type & IPV6_ADDR_MAPPED))
return -EINVAL;
/* It is a mapped address. Need to do some sanity
* checks.
*/
addr4 = sin6->sin6_addr.s6_addr32[3];
if (addr4 == htonl(INADDR_ANY) ||
addr4 == htonl(INADDR_BROADCAST) ||
ipv4_is_multicast(addr4))
return -EINVAL;
}
/* The scope ID must be specified for link local address. */
if (addr_type & IPV6_ADDR_LINKLOCAL) {
if (sin6->sin6_scope_id == 0)
return -EINVAL;
scope_id = sin6->sin6_scope_id;
}
binding_addr = &sin6->sin6_addr;
port = sin6->sin6_port;
#endif
} else {
return -EINVAL;
}
lock_sock(sk);
/* RDS socket does not allow re-binding. */
if (!ipv6_addr_any(&rs->rs_bound_addr)) {
ret = -EINVAL;
goto out;
}
/* Socket is connected. The binding address should have the same
* scope ID as the connected address, except the case when one is
* non-link local address (scope_id is 0).
*/
if (!ipv6_addr_any(&rs->rs_conn_addr) && scope_id &&
rs->rs_bound_scope_id &&
scope_id != rs->rs_bound_scope_id) {
ret = -EINVAL;
goto out;
}
/* The transport can be set using SO_RDS_TRANSPORT option before the
* socket is bound.
*/
if (rs->rs_transport) {
trans = rs->rs_transport;
if (!trans->laddr_check ||
trans->laddr_check(sock_net(sock->sk),
binding_addr, scope_id) != 0) {
ret = -ENOPROTOOPT;
goto out;
}
} else {
trans = rds_trans_get_preferred(sock_net(sock->sk),
binding_addr, scope_id);
if (!trans) {
ret = -EADDRNOTAVAIL;
pr_info_ratelimited("RDS: %s could not find a transport for %pI6c, load rds_tcp or rds_rdma?\n",
__func__, binding_addr);
goto out;
}
rs->rs_transport = trans;
}
sock_set_flag(sk, SOCK_RCU_FREE);
ret = rds_add_bound(rs, binding_addr, &port, scope_id);
if (ret)
rs->rs_transport = NULL;
out:
release_sock(sk);
return ret;
}
void rds_bind_lock_destroy(void)
{
rhashtable_destroy(&bind_hash_table);
}
int rds_bind_lock_init(void)
{
return rhashtable_init(&bind_hash_table, &ht_parms);
}