ipv6: remove hard coded limitation on ipv6_pinfo

IPv6 inet sockets are supposed to have a "struct ipv6_pinfo"
field at the end of their definition, so that inet6_sk_generic()
can derive from socket size the offset of the "struct ipv6_pinfo".

This is very fragile, and prevents adding bigger alignment
in sockets, because inet6_sk_generic() does not work
if the compiler adds padding after the ipv6_pinfo component.

We are currently working on a patch series to reorganize
TCP structures for better data locality and found issues
similar to the one fixed in commit f5d547676ca0
("tcp: fix tcp_inet6_sk() for 32bit kernels")

Alternative would be to force an alignment on "struct ipv6_pinfo",
greater or equal to __alignof__(any ipv6 sock) to ensure there is
no padding. This does not look great.

v2: fix typo in mptcp_proto_v6_init() (Paolo)

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Chao Wu <wwchao@google.com>
Cc: Wei Wang <weiwan@google.com>
Cc: Coco Li <lixiaoyan@google.com>
Cc: YiFei Zhu <zhuyifei@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 files changed, 7 insertions, 2 deletions

diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 5d593ddc0347..9f9c4b838664 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -102,9 +102,9 @@ bool ipv6_mod_enabled(void)
 }
 EXPORT_SYMBOL_GPL(ipv6_mod_enabled);
 
-static __inline__ struct ipv6_pinfo *inet6_sk_generic(struct sock *sk)
+static struct ipv6_pinfo *inet6_sk_generic(struct sock *sk)
 {
-	const int offset = sk->sk_prot->obj_size - sizeof(struct ipv6_pinfo);
+	const int offset = sk->sk_prot->ipv6_pinfo_offset;
 
 	return (struct ipv6_pinfo *)(((u8 *)sk) + offset);
 }
diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c
index f804c11e2146..2a0e8bc07398 100644
--- a/net/ipv6/ping.c
+++ b/net/ipv6/ping.c
@@ -215,6 +215,7 @@ struct proto pingv6_prot = {
 	.get_port =	ping_get_port,
 	.put_port =	ping_unhash,
 	.obj_size =	sizeof(struct raw6_sock),
+	.ipv6_pinfo_offset = offsetof(struct raw6_sock, inet6),
 };
 EXPORT_SYMBOL_GPL(pingv6_prot);
 
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index ac1cef094c5f..0fcf1b890807 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -1216,6 +1216,7 @@ struct proto rawv6_prot = {
 	.hash		   = raw_hash_sk,
 	.unhash		   = raw_unhash_sk,
 	.obj_size	   = sizeof(struct raw6_sock),
+	.ipv6_pinfo_offset = offsetof(struct raw6_sock, inet6),
 	.useroffset	   = offsetof(struct raw6_sock, filter),
 	.usersize	   = sizeof_field(struct raw6_sock, filter),
 	.h.raw_hash	   = &raw_v6_hashinfo,
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 4714eb695913..1b4529e833a1 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2175,6 +2175,7 @@ struct proto tcpv6_prot = {
 	.sysctl_rmem_offset	= offsetof(struct net, ipv4.sysctl_tcp_rmem),
 	.max_header		= MAX_TCP_HEADER,
 	.obj_size		= sizeof(struct tcp6_sock),
+	.ipv6_pinfo_offset = offsetof(struct tcp6_sock, inet6),
 	.slab_flags		= SLAB_TYPESAFE_BY_RCU,
 	.twsk_prot		= &tcp6_timewait_sock_ops,
 	.rsk_prot		= &tcp6_request_sock_ops,
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index b7c972aa09a7..95c75d8f73d5 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1798,6 +1798,7 @@ struct proto udpv6_prot = {
 	.sysctl_wmem_offset     = offsetof(struct net, ipv4.sysctl_udp_wmem_min),
 	.sysctl_rmem_offset     = offsetof(struct net, ipv4.sysctl_udp_rmem_min),
 	.obj_size		= sizeof(struct udp6_sock),
+	.ipv6_pinfo_offset = offsetof(struct udp6_sock, inet6),
 	.h.udp_table		= NULL,
 	.diag_destroy		= udp_abort,
 };
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 8e010d07917a..267d491e9707 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -67,6 +67,7 @@ struct proto udplitev6_prot = {
 	.sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_udp_wmem_min),
 	.sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min),
 	.obj_size	   = sizeof(struct udp6_sock),
+	.ipv6_pinfo_offset = offsetof(struct udp6_sock, inet6),
 	.h.udp_table	   = &udplite_table,
 };