| Message ID | 20240325181923.48769-2-kuniyu@amazon.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | tcp: Fix bind() regression and more tests. | expand |
On Mon, 25 Mar 2024 11:19:16 -0700 Kuniyuki Iwashima wrote: > - if (sk->sk_family == AF_INET && ipv6_only_sock(sk2)) > + if (ipv6_only_sock(sk2) && > + (sk->sk_family == AF_INET || ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr))) breaks build for IPV6=n (i.e. TDC): https://github.com/p4tc-dev/tc-executor/commit/ac16181d7589bdf29c7c3907243e829f6b954570#diff-0042f6af11ac801c4370fb95b8e4f0de734b793c9e23bff936a1bb03c63eb6f0R228
From: Jakub Kicinski <kuba@kernel.org> Date: Mon, 25 Mar 2024 16:56:33 -0700 > On Mon, 25 Mar 2024 11:19:16 -0700 Kuniyuki Iwashima wrote: > > - if (sk->sk_family == AF_INET && ipv6_only_sock(sk2)) > > + if (ipv6_only_sock(sk2) && > > + (sk->sk_family == AF_INET || ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr))) > > breaks build for IPV6=n (i.e. TDC): > > https://github.com/p4tc-dev/tc-executor/commit/ac16181d7589bdf29c7c3907243e829f6b954570#diff-0042f6af11ac801c4370fb95b8e4f0de734b793c9e23bff936a1bb03c63eb6f0R228 Oops, will fix it in v2. Thanks!
Hi Kuniyuki,
kernel test robot noticed the following build errors:
[auto build test ERROR on net/main]
url: https://github.com/intel-lab-lkp/linux/commits/Kuniyuki-Iwashima/tcp-Fix-bind-regression-for-v6-only-wildcard-and-v4-mapped-v6-non-wildcard-addresses/20240326-024257
base: net/main
patch link: https://lore.kernel.org/r/20240325181923.48769-2-kuniyu%40amazon.com
patch subject: [PATCH v1 net 1/8] tcp: Fix bind() regression for v6-only wildcard and v4-mapped-v6 non-wildcard addresses.
config: alpha-defconfig (https://download.01.org/0day-ci/archive/20240326/202403261847.wj7EwLat-lkp@intel.com/config)
compiler: alpha-linux-gcc (GCC) 13.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240326/202403261847.wj7EwLat-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202403261847.wj7EwLat-lkp@intel.com/
All errors (new ones prefixed by >>):
In file included from include/net/inet_sock.h:22,
from include/net/inet_connection_sock.h:21,
from net/ipv4/inet_connection_sock.c:15:
net/ipv4/inet_connection_sock.c: In function '__inet_bhash2_conflict':
>> include/net/sock.h:375:37: error: 'const struct sock_common' has no member named 'skc_v6_rcv_saddr'; did you mean 'skc_rcv_saddr'?
375 | #define sk_v6_rcv_saddr __sk_common.skc_v6_rcv_saddr
| ^~~~~~~~~~~~~~~~
net/ipv4/inet_connection_sock.c:207:66: note: in expansion of macro 'sk_v6_rcv_saddr'
207 | (sk->sk_family == AF_INET || ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr)))
| ^~~~~~~~~~~~~~~
vim +375 include/net/sock.h
4dc6dc7162c08b Eric Dumazet 2009-07-15 354
68835aba4d9b74 Eric Dumazet 2010-11-30 355 #define sk_dontcopy_begin __sk_common.skc_dontcopy_begin
68835aba4d9b74 Eric Dumazet 2010-11-30 356 #define sk_dontcopy_end __sk_common.skc_dontcopy_end
4dc6dc7162c08b Eric Dumazet 2009-07-15 357 #define sk_hash __sk_common.skc_hash
5080546682bae3 Eric Dumazet 2013-10-02 358 #define sk_portpair __sk_common.skc_portpair
05dbc7b59481ca Eric Dumazet 2013-10-03 359 #define sk_num __sk_common.skc_num
05dbc7b59481ca Eric Dumazet 2013-10-03 360 #define sk_dport __sk_common.skc_dport
5080546682bae3 Eric Dumazet 2013-10-02 361 #define sk_addrpair __sk_common.skc_addrpair
5080546682bae3 Eric Dumazet 2013-10-02 362 #define sk_daddr __sk_common.skc_daddr
5080546682bae3 Eric Dumazet 2013-10-02 363 #define sk_rcv_saddr __sk_common.skc_rcv_saddr
^1da177e4c3f41 Linus Torvalds 2005-04-16 364 #define sk_family __sk_common.skc_family
^1da177e4c3f41 Linus Torvalds 2005-04-16 365 #define sk_state __sk_common.skc_state
^1da177e4c3f41 Linus Torvalds 2005-04-16 366 #define sk_reuse __sk_common.skc_reuse
055dc21a1d1d21 Tom Herbert 2013-01-22 367 #define sk_reuseport __sk_common.skc_reuseport
9fe516ba3fb29b Eric Dumazet 2014-06-27 368 #define sk_ipv6only __sk_common.skc_ipv6only
26abe14379f8e2 Eric W. Biederman 2015-05-08 369 #define sk_net_refcnt __sk_common.skc_net_refcnt
^1da177e4c3f41 Linus Torvalds 2005-04-16 370 #define sk_bound_dev_if __sk_common.skc_bound_dev_if
^1da177e4c3f41 Linus Torvalds 2005-04-16 371 #define sk_bind_node __sk_common.skc_bind_node
8feaf0c0a5488b Arnaldo Carvalho de Melo 2005-08-09 372 #define sk_prot __sk_common.skc_prot
07feaebfcc10cd Eric W. Biederman 2007-09-12 373 #define sk_net __sk_common.skc_net
efe4208f47f907 Eric Dumazet 2013-10-03 374 #define sk_v6_daddr __sk_common.skc_v6_daddr
efe4208f47f907 Eric Dumazet 2013-10-03 @375 #define sk_v6_rcv_saddr __sk_common.skc_v6_rcv_saddr
33cf7c90fe2f97 Eric Dumazet 2015-03-11 376 #define sk_cookie __sk_common.skc_cookie
70da268b569d32 Eric Dumazet 2015-10-08 377 #define sk_incoming_cpu __sk_common.skc_incoming_cpu
8e5eb54d303b7c Eric Dumazet 2015-10-08 378 #define sk_flags __sk_common.skc_flags
ed53d0ab761f5c Eric Dumazet 2015-10-08 379 #define sk_rxhash __sk_common.skc_rxhash
efe4208f47f907 Eric Dumazet 2013-10-03 380
5d4cc87414c5d1 Eric Dumazet 2024-02-16 381 __cacheline_group_begin(sock_write_rx);
43f51df4172955 Eric Dumazet 2021-11-15 382
9115e8cd2a0c6e Eric Dumazet 2016-12-03 383 atomic_t sk_drops;
5d4cc87414c5d1 Eric Dumazet 2024-02-16 384 __s32 sk_peek_off;
9115e8cd2a0c6e Eric Dumazet 2016-12-03 385 struct sk_buff_head sk_error_queue;
b178bb3dfc30d9 Eric Dumazet 2010-11-16 386 struct sk_buff_head sk_receive_queue;
fa438ccfdfd3f6 Eric Dumazet 2007-03-04 387 /*
fa438ccfdfd3f6 Eric Dumazet 2007-03-04 388 * The backlog queue is special, it is always used with
fa438ccfdfd3f6 Eric Dumazet 2007-03-04 389 * the per-socket spinlock held and requires low latency
fa438ccfdfd3f6 Eric Dumazet 2007-03-04 390 * access. Therefore we special case it's implementation.
b178bb3dfc30d9 Eric Dumazet 2010-11-16 391 * Note : rmem_alloc is in this structure to fill a hole
b178bb3dfc30d9 Eric Dumazet 2010-11-16 392 * on 64bit arches, not because its logically part of
b178bb3dfc30d9 Eric Dumazet 2010-11-16 393 * backlog.
fa438ccfdfd3f6 Eric Dumazet 2007-03-04 394 */
fa438ccfdfd3f6 Eric Dumazet 2007-03-04 395 struct {
b178bb3dfc30d9 Eric Dumazet 2010-11-16 396 atomic_t rmem_alloc;
b178bb3dfc30d9 Eric Dumazet 2010-11-16 397 int len;
fa438ccfdfd3f6 Eric Dumazet 2007-03-04 398 struct sk_buff *head;
fa438ccfdfd3f6 Eric Dumazet 2007-03-04 399 struct sk_buff *tail;
fa438ccfdfd3f6 Eric Dumazet 2007-03-04 400 } sk_backlog;
b178bb3dfc30d9 Eric Dumazet 2010-11-16 401 #define sk_rmem_alloc sk_backlog.rmem_alloc
2c8c56e15df3d4 Eric Dumazet 2014-11-11 402
5d4cc87414c5d1 Eric Dumazet 2024-02-16 403 __cacheline_group_end(sock_write_rx);
5d4cc87414c5d1 Eric Dumazet 2024-02-16 404
5d4cc87414c5d1 Eric Dumazet 2024-02-16 405 __cacheline_group_begin(sock_read_rx);
5d4cc87414c5d1 Eric Dumazet 2024-02-16 406 /* early demux fields */
5d4cc87414c5d1 Eric Dumazet 2024-02-16 407 struct dst_entry __rcu *sk_rx_dst;
5d4cc87414c5d1 Eric Dumazet 2024-02-16 408 int sk_rx_dst_ifindex;
5d4cc87414c5d1 Eric Dumazet 2024-02-16 409 u32 sk_rx_dst_cookie;
5d4cc87414c5d1 Eric Dumazet 2024-02-16 410
Hi Kuniyuki, kernel test robot noticed the following build errors: [auto build test ERROR on net/main] url: https://github.com/intel-lab-lkp/linux/commits/Kuniyuki-Iwashima/tcp-Fix-bind-regression-for-v6-only-wildcard-and-v4-mapped-v6-non-wildcard-addresses/20240326-024257 base: net/main patch link: https://lore.kernel.org/r/20240325181923.48769-2-kuniyu%40amazon.com patch subject: [PATCH v1 net 1/8] tcp: Fix bind() regression for v6-only wildcard and v4-mapped-v6 non-wildcard addresses. config: x86_64-kexec (https://download.01.org/0day-ci/archive/20240326/202403261821.5fkObY55-lkp@intel.com/config) compiler: clang version 17.0.6 (https://github.com/llvm/llvm-project 6009708b4367171ccdbf4b5905cb6a803753fe18) reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240326/202403261821.5fkObY55-lkp@intel.com/reproduce) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp@intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202403261821.5fkObY55-lkp@intel.com/ All errors (new ones prefixed by >>): >> net/ipv4/inet_connection_sock.c:207:59: error: no member named 'skc_v6_rcv_saddr' in 'struct sock_common'; did you mean 'skc_rcv_saddr'? 207 | (sk->sk_family == AF_INET || ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr))) | ^ include/net/sock.h:375:37: note: expanded from macro 'sk_v6_rcv_saddr' 375 | #define sk_v6_rcv_saddr __sk_common.skc_v6_rcv_saddr | ^ include/net/sock.h:155:11: note: 'skc_rcv_saddr' declared here 155 | __be32 skc_rcv_saddr; | ^ 1 error generated. vim +207 net/ipv4/inet_connection_sock.c 201 202 static bool __inet_bhash2_conflict(const struct sock *sk, struct sock *sk2, 203 kuid_t sk_uid, bool relax, 204 bool reuseport_cb_ok, bool reuseport_ok) 205 { 206 if (ipv6_only_sock(sk2) && > 207 (sk->sk_family == AF_INET || ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr))) 208 return false; 209 210 return inet_bind_conflict(sk, sk2, sk_uid, relax, 211 reuseport_cb_ok, reuseport_ok); 212 } 213
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 7d8090f109ef..612aa1d2eff7 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -203,7 +203,8 @@ static bool __inet_bhash2_conflict(const struct sock *sk, struct sock *sk2, kuid_t sk_uid, bool relax, bool reuseport_cb_ok, bool reuseport_ok) { - if (sk->sk_family == AF_INET && ipv6_only_sock(sk2)) + if (ipv6_only_sock(sk2) && + (sk->sk_family == AF_INET || ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr))) return false; return inet_bind_conflict(sk, sk2, sk_uid, relax,
Commit 5e07e672412b ("tcp: Use bhash2 for v4-mapped-v6 non-wildcard address.") introduced bind() regression for v4-mapped-v6 address. When we bind() the following two addresses on the same port, the 2nd bind() should succeed but fails now. 1. [::] w/ IPV6_ONLY 2. ::ffff:127.0.0.1 After the chagne, v4-mapped-v6 uses bhash2 instead of bhash to detect conflict faster, but I forgot to add a necessary change. During the 2nd bind(), inet_bind2_bucket_match_addr_any() returns the tb2 bucket of [::], and inet_bhash2_conflict() finally calls inet_bind_conflict(), which returns true, meaning conflict. inet_bhash2_addr_any_conflict |- inet_bind2_bucket_match_addr_any <-- return [::] bucket `- inet_bhash2_conflict `- __inet_bhash2_conflict <-- checks IPV6_ONLY for AF_INET | but not for v4-mapped-v6 address `- inet_bind_conflict <-- does not check address inet_bind_conflict() does not check socket addresses because __inet_bhash2_conflict() is expected to do so. However, it checks IPV6_V6ONLY attribute only against AF_INET socket, and not for v4-mapped-v6 address. As a result, v4-mapped-v6 address conflicts with v6-only wildcard address. To avoid that, let's add the missing test to use bhash2 for v4-mapped-v6 address. Fixes: 5e07e672412b ("tcp: Use bhash2 for v4-mapped-v6 non-wildcard address.") Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> --- net/ipv4/inet_connection_sock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)