Commit 26333c37 authored by Sabrina Dubroca's avatar Sabrina Dubroca Committed by Steffen Klassert

xfrm: add IPv6 support for espintcp

This extends espintcp to support IPv6, building on the existing code
and the new UDPv6 encapsulation support. Most of the code is either
reused directly (stream parser, ULP) or very similar to the IPv4
variant (net/ipv6/esp6.c changes).

The separation of config options for IPv4 and IPv6 espintcp requires a
bit of Kconfig gymnastics to enable the core code.
Signed-off-by: default avatarSabrina Dubroca <sd@queasysnail.net>
Signed-off-by: default avatarSteffen Klassert <steffen.klassert@secunet.com>
parent 0146dca7
...@@ -58,6 +58,8 @@ struct ipv6_stub { ...@@ -58,6 +58,8 @@ struct ipv6_stub {
bool router, bool solicited, bool override, bool inc_opt); bool router, bool solicited, bool override, bool inc_opt);
#if IS_ENABLED(CONFIG_XFRM) #if IS_ENABLED(CONFIG_XFRM)
int (*xfrm6_udp_encap_rcv)(struct sock *sk, struct sk_buff *skb); int (*xfrm6_udp_encap_rcv)(struct sock *sk, struct sk_buff *skb);
int (*xfrm6_rcv_encap)(struct sk_buff *skb, int nexthdr, __be32 spi,
int encap_type);
#endif #endif
struct neigh_table *nd_tbl; struct neigh_table *nd_tbl;
}; };
......
...@@ -384,6 +384,7 @@ config INET_ESPINTCP ...@@ -384,6 +384,7 @@ config INET_ESPINTCP
depends on XFRM && INET_ESP depends on XFRM && INET_ESP
select STREAM_PARSER select STREAM_PARSER
select NET_SOCK_MSG select NET_SOCK_MSG
select XFRM_ESPINTCP
help help
Support for RFC 8229 encapsulation of ESP and IKE over Support for RFC 8229 encapsulation of ESP and IKE over
TCP/IPv4 sockets. TCP/IPv4 sockets.
......
...@@ -88,6 +88,18 @@ config INET6_ESP_OFFLOAD ...@@ -88,6 +88,18 @@ config INET6_ESP_OFFLOAD
If unsure, say N. If unsure, say N.
config INET6_ESPINTCP
bool "IPv6: ESP in TCP encapsulation (RFC 8229)"
depends on XFRM && INET6_ESP
select STREAM_PARSER
select NET_SOCK_MSG
select XFRM_ESPINTCP
help
Support for RFC 8229 encapsulation of ESP and IKE over
TCP/IPv6 sockets.
If unsure, say N.
config INET6_IPCOMP config INET6_IPCOMP
tristate "IPv6: IPComp transformation" tristate "IPv6: IPComp transformation"
select INET6_XFRM_TUNNEL select INET6_XFRM_TUNNEL
......
...@@ -964,6 +964,7 @@ static const struct ipv6_stub ipv6_stub_impl = { ...@@ -964,6 +964,7 @@ static const struct ipv6_stub ipv6_stub_impl = {
.ndisc_send_na = ndisc_send_na, .ndisc_send_na = ndisc_send_na,
#if IS_ENABLED(CONFIG_XFRM) #if IS_ENABLED(CONFIG_XFRM)
.xfrm6_udp_encap_rcv = xfrm6_udp_encap_rcv, .xfrm6_udp_encap_rcv = xfrm6_udp_encap_rcv,
.xfrm6_rcv_encap = xfrm6_rcv_encap,
#endif #endif
.nd_tbl = &nd_tbl, .nd_tbl = &nd_tbl,
}; };
......
...@@ -33,6 +33,9 @@ ...@@ -33,6 +33,9 @@
#include <net/protocol.h> #include <net/protocol.h>
#include <net/udp.h> #include <net/udp.h>
#include <linux/icmpv6.h> #include <linux/icmpv6.h>
#include <net/tcp.h>
#include <net/espintcp.h>
#include <net/inet6_hashtables.h>
#include <linux/highmem.h> #include <linux/highmem.h>
...@@ -132,6 +135,132 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp) ...@@ -132,6 +135,132 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp)
put_page(sg_page(sg)); put_page(sg_page(sg));
} }
#ifdef CONFIG_INET6_ESPINTCP
struct esp_tcp_sk {
struct sock *sk;
struct rcu_head rcu;
};
static void esp_free_tcp_sk(struct rcu_head *head)
{
struct esp_tcp_sk *esk = container_of(head, struct esp_tcp_sk, rcu);
sock_put(esk->sk);
kfree(esk);
}
static struct sock *esp6_find_tcp_sk(struct xfrm_state *x)
{
struct xfrm_encap_tmpl *encap = x->encap;
struct esp_tcp_sk *esk;
__be16 sport, dport;
struct sock *nsk;
struct sock *sk;
sk = rcu_dereference(x->encap_sk);
if (sk && sk->sk_state == TCP_ESTABLISHED)
return sk;
spin_lock_bh(&x->lock);
sport = encap->encap_sport;
dport = encap->encap_dport;
nsk = rcu_dereference_protected(x->encap_sk,
lockdep_is_held(&x->lock));
if (sk && sk == nsk) {
esk = kmalloc(sizeof(*esk), GFP_ATOMIC);
if (!esk) {
spin_unlock_bh(&x->lock);
return ERR_PTR(-ENOMEM);
}
RCU_INIT_POINTER(x->encap_sk, NULL);
esk->sk = sk;
call_rcu(&esk->rcu, esp_free_tcp_sk);
}
spin_unlock_bh(&x->lock);
sk = __inet6_lookup_established(xs_net(x), &tcp_hashinfo, &x->id.daddr.in6,
dport, &x->props.saddr.in6, ntohs(sport), 0, 0);
if (!sk)
return ERR_PTR(-ENOENT);
if (!tcp_is_ulp_esp(sk)) {
sock_put(sk);
return ERR_PTR(-EINVAL);
}
spin_lock_bh(&x->lock);
nsk = rcu_dereference_protected(x->encap_sk,
lockdep_is_held(&x->lock));
if (encap->encap_sport != sport ||
encap->encap_dport != dport) {
sock_put(sk);
sk = nsk ?: ERR_PTR(-EREMCHG);
} else if (sk == nsk) {
sock_put(sk);
} else {
rcu_assign_pointer(x->encap_sk, sk);
}
spin_unlock_bh(&x->lock);
return sk;
}
static int esp_output_tcp_finish(struct xfrm_state *x, struct sk_buff *skb)
{
struct sock *sk;
int err;
rcu_read_lock();
sk = esp6_find_tcp_sk(x);
err = PTR_ERR_OR_ZERO(sk);
if (err)
goto out;
bh_lock_sock(sk);
if (sock_owned_by_user(sk))
err = espintcp_queue_out(sk, skb);
else
err = espintcp_push_skb(sk, skb);
bh_unlock_sock(sk);
out:
rcu_read_unlock();
return err;
}
static int esp_output_tcp_encap_cb(struct net *net, struct sock *sk,
struct sk_buff *skb)
{
struct dst_entry *dst = skb_dst(skb);
struct xfrm_state *x = dst->xfrm;
return esp_output_tcp_finish(x, skb);
}
static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
{
int err;
local_bh_disable();
err = xfrm_trans_queue_net(xs_net(x), skb, esp_output_tcp_encap_cb);
local_bh_enable();
/* EINPROGRESS just happens to do the right thing. It
* actually means that the skb has been consumed and
* isn't coming back.
*/
return err ?: -EINPROGRESS;
}
#else
static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
{
kfree_skb(skb);
return -EOPNOTSUPP;
}
#endif
static void esp_output_encap_csum(struct sk_buff *skb) static void esp_output_encap_csum(struct sk_buff *skb)
{ {
/* UDP encap with IPv6 requires a valid checksum */ /* UDP encap with IPv6 requires a valid checksum */
...@@ -181,6 +310,10 @@ static void esp_output_done(struct crypto_async_request *base, int err) ...@@ -181,6 +310,10 @@ static void esp_output_done(struct crypto_async_request *base, int err)
secpath_reset(skb); secpath_reset(skb);
xfrm_dev_resume(skb); xfrm_dev_resume(skb);
} else { } else {
if (!err &&
x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
esp_output_tail_tcp(x, skb);
else
xfrm_output_resume(skb, err); xfrm_output_resume(skb, err);
} }
} }
...@@ -274,6 +407,41 @@ static struct ip_esp_hdr *esp6_output_udp_encap(struct sk_buff *skb, ...@@ -274,6 +407,41 @@ static struct ip_esp_hdr *esp6_output_udp_encap(struct sk_buff *skb,
return (struct ip_esp_hdr *)(uh + 1); return (struct ip_esp_hdr *)(uh + 1);
} }
#ifdef CONFIG_INET6_ESPINTCP
static struct ip_esp_hdr *esp6_output_tcp_encap(struct xfrm_state *x,
struct sk_buff *skb,
struct esp_info *esp)
{
__be16 *lenp = (void *)esp->esph;
struct ip_esp_hdr *esph;
unsigned int len;
struct sock *sk;
len = skb->len + esp->tailen - skb_transport_offset(skb);
if (len > IP_MAX_MTU)
return ERR_PTR(-EMSGSIZE);
rcu_read_lock();
sk = esp6_find_tcp_sk(x);
rcu_read_unlock();
if (IS_ERR(sk))
return ERR_CAST(sk);
*lenp = htons(len);
esph = (struct ip_esp_hdr *)(lenp + 1);
return esph;
}
#else
static struct ip_esp_hdr *esp6_output_tcp_encap(struct xfrm_state *x,
struct sk_buff *skb,
struct esp_info *esp)
{
return ERR_PTR(-EOPNOTSUPP);
}
#endif
static int esp6_output_encap(struct xfrm_state *x, struct sk_buff *skb, static int esp6_output_encap(struct xfrm_state *x, struct sk_buff *skb,
struct esp_info *esp) struct esp_info *esp)
{ {
...@@ -294,6 +462,9 @@ static int esp6_output_encap(struct xfrm_state *x, struct sk_buff *skb, ...@@ -294,6 +462,9 @@ static int esp6_output_encap(struct xfrm_state *x, struct sk_buff *skb,
case UDP_ENCAP_ESPINUDP_NON_IKE: case UDP_ENCAP_ESPINUDP_NON_IKE:
esph = esp6_output_udp_encap(skb, encap_type, esp, sport, dport); esph = esp6_output_udp_encap(skb, encap_type, esp, sport, dport);
break; break;
case TCP_ENCAP_ESPINTCP:
esph = esp6_output_tcp_encap(x, skb, esp);
break;
} }
if (IS_ERR(esph)) if (IS_ERR(esph))
...@@ -509,6 +680,9 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info ...@@ -509,6 +680,9 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info
if (sg != dsg) if (sg != dsg)
esp_ssg_unref(x, tmp); esp_ssg_unref(x, tmp);
if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
err = esp_output_tail_tcp(x, skb);
error_free: error_free:
kfree(tmp); kfree(tmp);
error: error:
...@@ -633,9 +807,13 @@ int esp6_input_done2(struct sk_buff *skb, int err) ...@@ -633,9 +807,13 @@ int esp6_input_done2(struct sk_buff *skb, int err)
const struct ipv6hdr *ip6h = ipv6_hdr(skb); const struct ipv6hdr *ip6h = ipv6_hdr(skb);
struct xfrm_encap_tmpl *encap = x->encap; struct xfrm_encap_tmpl *encap = x->encap;
struct udphdr *uh = (void *)(skb_network_header(skb) + hdr_len); struct udphdr *uh = (void *)(skb_network_header(skb) + hdr_len);
struct tcphdr *th = (void *)(skb_network_header(skb) + hdr_len);
__be16 source; __be16 source;
switch (x->encap->encap_type) { switch (x->encap->encap_type) {
case TCP_ENCAP_ESPINTCP:
source = th->source;
break;
case UDP_ENCAP_ESPINUDP: case UDP_ENCAP_ESPINUDP:
case UDP_ENCAP_ESPINUDP_NON_IKE: case UDP_ENCAP_ESPINUDP_NON_IKE:
source = uh->source; source = uh->source;
...@@ -1039,6 +1217,14 @@ static int esp6_init_state(struct xfrm_state *x) ...@@ -1039,6 +1217,14 @@ static int esp6_init_state(struct xfrm_state *x)
case UDP_ENCAP_ESPINUDP_NON_IKE: case UDP_ENCAP_ESPINUDP_NON_IKE:
x->props.header_len += sizeof(struct udphdr) + 2 * sizeof(u32); x->props.header_len += sizeof(struct udphdr) + 2 * sizeof(u32);
break; break;
#ifdef CONFIG_INET6_ESPINTCP
case TCP_ENCAP_ESPINTCP:
/* only the length field, TCP encap is done by
* the socket
*/
x->props.header_len += 2;
break;
#endif
} }
} }
......
...@@ -99,4 +99,7 @@ config NET_KEY_MIGRATE ...@@ -99,4 +99,7 @@ config NET_KEY_MIGRATE
If unsure, say N. If unsure, say N.
config XFRM_ESPINTCP
bool
endif # INET endif # INET
...@@ -11,4 +11,4 @@ obj-$(CONFIG_XFRM_ALGO) += xfrm_algo.o ...@@ -11,4 +11,4 @@ obj-$(CONFIG_XFRM_ALGO) += xfrm_algo.o
obj-$(CONFIG_XFRM_USER) += xfrm_user.o obj-$(CONFIG_XFRM_USER) += xfrm_user.o
obj-$(CONFIG_XFRM_IPCOMP) += xfrm_ipcomp.o obj-$(CONFIG_XFRM_IPCOMP) += xfrm_ipcomp.o
obj-$(CONFIG_XFRM_INTERFACE) += xfrm_interface.o obj-$(CONFIG_XFRM_INTERFACE) += xfrm_interface.o
obj-$(CONFIG_INET_ESPINTCP) += espintcp.o obj-$(CONFIG_XFRM_ESPINTCP) += espintcp.o
...@@ -6,6 +6,9 @@ ...@@ -6,6 +6,9 @@
#include <net/espintcp.h> #include <net/espintcp.h>
#include <linux/skmsg.h> #include <linux/skmsg.h>
#include <net/inet_common.h> #include <net/inet_common.h>
#if IS_ENABLED(CONFIG_IPV6)
#include <net/ipv6_stubs.h>
#endif
static void handle_nonesp(struct espintcp_ctx *ctx, struct sk_buff *skb, static void handle_nonesp(struct espintcp_ctx *ctx, struct sk_buff *skb,
struct sock *sk) struct sock *sk)
...@@ -31,6 +34,11 @@ static void handle_esp(struct sk_buff *skb, struct sock *sk) ...@@ -31,6 +34,11 @@ static void handle_esp(struct sk_buff *skb, struct sock *sk)
rcu_read_lock(); rcu_read_lock();
skb->dev = dev_get_by_index_rcu(sock_net(sk), skb->skb_iif); skb->dev = dev_get_by_index_rcu(sock_net(sk), skb->skb_iif);
local_bh_disable(); local_bh_disable();
#if IS_ENABLED(CONFIG_IPV6)
if (sk->sk_family == AF_INET6)
ipv6_stub->xfrm6_rcv_encap(skb, IPPROTO_ESP, 0, TCP_ENCAP_ESPINTCP);
else
#endif
xfrm4_rcv_encap(skb, IPPROTO_ESP, 0, TCP_ENCAP_ESPINTCP); xfrm4_rcv_encap(skb, IPPROTO_ESP, 0, TCP_ENCAP_ESPINTCP);
local_bh_enable(); local_bh_enable();
rcu_read_unlock(); rcu_read_unlock();
...@@ -347,6 +355,9 @@ static int espintcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) ...@@ -347,6 +355,9 @@ static int espintcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
static struct proto espintcp_prot __ro_after_init; static struct proto espintcp_prot __ro_after_init;
static struct proto_ops espintcp_ops __ro_after_init; static struct proto_ops espintcp_ops __ro_after_init;
static struct proto espintcp6_prot;
static struct proto_ops espintcp6_ops;
static DEFINE_MUTEX(tcpv6_prot_mutex);
static void espintcp_data_ready(struct sock *sk) static void espintcp_data_ready(struct sock *sk)
{ {
...@@ -384,10 +395,14 @@ static void espintcp_destruct(struct sock *sk) ...@@ -384,10 +395,14 @@ static void espintcp_destruct(struct sock *sk)
bool tcp_is_ulp_esp(struct sock *sk) bool tcp_is_ulp_esp(struct sock *sk)
{ {
return sk->sk_prot == &espintcp_prot; return sk->sk_prot == &espintcp_prot || sk->sk_prot == &espintcp6_prot;
} }
EXPORT_SYMBOL_GPL(tcp_is_ulp_esp); EXPORT_SYMBOL_GPL(tcp_is_ulp_esp);
static void build_protos(struct proto *espintcp_prot,
struct proto_ops *espintcp_ops,
const struct proto *orig_prot,
const struct proto_ops *orig_ops);
static int espintcp_init_sk(struct sock *sk) static int espintcp_init_sk(struct sock *sk)
{ {
struct inet_connection_sock *icsk = inet_csk(sk); struct inet_connection_sock *icsk = inet_csk(sk);
...@@ -415,8 +430,19 @@ static int espintcp_init_sk(struct sock *sk) ...@@ -415,8 +430,19 @@ static int espintcp_init_sk(struct sock *sk)
strp_check_rcv(&ctx->strp); strp_check_rcv(&ctx->strp);
skb_queue_head_init(&ctx->ike_queue); skb_queue_head_init(&ctx->ike_queue);
skb_queue_head_init(&ctx->out_queue); skb_queue_head_init(&ctx->out_queue);
if (sk->sk_family == AF_INET) {
sk->sk_prot = &espintcp_prot; sk->sk_prot = &espintcp_prot;
sk->sk_socket->ops = &espintcp_ops; sk->sk_socket->ops = &espintcp_ops;
} else {
mutex_lock(&tcpv6_prot_mutex);
if (!espintcp6_prot.recvmsg)
build_protos(&espintcp6_prot, &espintcp6_ops, sk->sk_prot, sk->sk_socket->ops);
mutex_unlock(&tcpv6_prot_mutex);
sk->sk_prot = &espintcp6_prot;
sk->sk_socket->ops = &espintcp6_ops;
}
ctx->saved_data_ready = sk->sk_data_ready; ctx->saved_data_ready = sk->sk_data_ready;
ctx->saved_write_space = sk->sk_write_space; ctx->saved_write_space = sk->sk_write_space;
sk->sk_data_ready = espintcp_data_ready; sk->sk_data_ready = espintcp_data_ready;
...@@ -489,6 +515,20 @@ static __poll_t espintcp_poll(struct file *file, struct socket *sock, ...@@ -489,6 +515,20 @@ static __poll_t espintcp_poll(struct file *file, struct socket *sock,
return mask; return mask;
} }
static void build_protos(struct proto *espintcp_prot,
struct proto_ops *espintcp_ops,
const struct proto *orig_prot,
const struct proto_ops *orig_ops)
{
memcpy(espintcp_prot, orig_prot, sizeof(struct proto));
memcpy(espintcp_ops, orig_ops, sizeof(struct proto_ops));
espintcp_prot->sendmsg = espintcp_sendmsg;
espintcp_prot->recvmsg = espintcp_recvmsg;
espintcp_prot->close = espintcp_close;
espintcp_prot->release_cb = espintcp_release;
espintcp_ops->poll = espintcp_poll;
}
static struct tcp_ulp_ops espintcp_ulp __read_mostly = { static struct tcp_ulp_ops espintcp_ulp __read_mostly = {
.name = "espintcp", .name = "espintcp",
.owner = THIS_MODULE, .owner = THIS_MODULE,
...@@ -497,13 +537,7 @@ static struct tcp_ulp_ops espintcp_ulp __read_mostly = { ...@@ -497,13 +537,7 @@ static struct tcp_ulp_ops espintcp_ulp __read_mostly = {
void __init espintcp_init(void) void __init espintcp_init(void)
{ {
memcpy(&espintcp_prot, &tcp_prot, sizeof(tcp_prot)); build_protos(&espintcp_prot, &espintcp_ops, &tcp_prot, &inet_stream_ops);
memcpy(&espintcp_ops, &inet_stream_ops, sizeof(inet_stream_ops));
espintcp_prot.sendmsg = espintcp_sendmsg;
espintcp_prot.recvmsg = espintcp_recvmsg;
espintcp_prot.close = espintcp_close;
espintcp_prot.release_cb = espintcp_release;
espintcp_ops.poll = espintcp_poll;
tcp_register_ulp(&espintcp_ulp); tcp_register_ulp(&espintcp_ulp);
} }
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment