Commit f296234c authored by Peter Krystad's avatar Peter Krystad Committed by David S. Miller

mptcp: Add handling of incoming MP_JOIN requests

Process the MP_JOIN option in a SYN packet with the same flow
as MP_CAPABLE but when the third ACK is received add the
subflow to the MPTCP socket subflow list instead of adding it to
the TCP socket accept queue.

The subflow is added at the end of the subflow list so it will not
interfere with the existing subflows operation and no data is
expected to be transmitted on it.
Co-developed-by: default avatarFlorian Westphal <fw@strlen.de>
Signed-off-by: default avatarFlorian Westphal <fw@strlen.de>
Co-developed-by: default avatarPaolo Abeni <pabeni@redhat.com>
Signed-off-by: default avatarPaolo Abeni <pabeni@redhat.com>
Signed-off-by: default avatarPeter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: default avatarMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 1b1c7a0e
...@@ -92,7 +92,13 @@ struct mptcp_options_received { ...@@ -92,7 +92,13 @@ struct mptcp_options_received {
add_addr : 1, add_addr : 1,
rm_addr : 1, rm_addr : 1,
family : 4, family : 4,
echo : 1; echo : 1,
backup : 1;
u32 token;
u32 nonce;
u64 thmac;
u8 hmac[20];
u8 join_id;
u8 use_map:1, u8 use_map:1,
dsn64:1, dsn64:1,
data_fin:1, data_fin:1,
......
...@@ -42,6 +42,10 @@ struct mptcp_out_options { ...@@ -42,6 +42,10 @@ struct mptcp_out_options {
u8 addr_id; u8 addr_id;
u64 ahmac; u64 ahmac;
u8 rm_id; u8 rm_id;
u8 join_id;
u8 backup;
u32 nonce;
u64 thmac;
struct mptcp_ext ext_copy; struct mptcp_ext ext_copy;
#endif #endif
}; };
...@@ -115,6 +119,8 @@ static inline bool mptcp_skb_can_collapse(const struct sk_buff *to, ...@@ -115,6 +119,8 @@ static inline bool mptcp_skb_can_collapse(const struct sk_buff *to,
skb_ext_find(from, SKB_EXT_MPTCP)); skb_ext_find(from, SKB_EXT_MPTCP));
} }
bool mptcp_sk_is_subflow(const struct sock *sk);
#else #else
static inline void mptcp_init(void) static inline void mptcp_init(void)
...@@ -181,6 +187,11 @@ static inline bool mptcp_skb_can_collapse(const struct sk_buff *to, ...@@ -181,6 +187,11 @@ static inline bool mptcp_skb_can_collapse(const struct sk_buff *to,
return true; return true;
} }
static inline bool mptcp_sk_is_subflow(const struct sock *sk)
{
return false;
}
#endif /* CONFIG_MPTCP */ #endif /* CONFIG_MPTCP */
#if IS_ENABLED(CONFIG_MPTCP_IPV6) #if IS_ENABLED(CONFIG_MPTCP_IPV6)
......
...@@ -774,6 +774,12 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, ...@@ -774,6 +774,12 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb,
if (!child) if (!child)
goto listen_overflow; goto listen_overflow;
if (own_req && sk_is_mptcp(child) && mptcp_sk_is_subflow(child)) {
reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req);
inet_csk_reqsk_queue_drop_and_put(sk, req);
return child;
}
sock_rps_save_rxhash(child, skb); sock_rps_save_rxhash(child, skb);
tcp_synack_rtt_meas(child, req); tcp_synack_rtt_meas(child, req);
*req_stolen = !own_req; *req_stolen = !own_req;
......
...@@ -96,6 +96,38 @@ void mptcp_parse_option(const struct sk_buff *skb, const unsigned char *ptr, ...@@ -96,6 +96,38 @@ void mptcp_parse_option(const struct sk_buff *skb, const unsigned char *ptr,
mp_opt->rcvr_key, mp_opt->data_len); mp_opt->rcvr_key, mp_opt->data_len);
break; break;
case MPTCPOPT_MP_JOIN:
mp_opt->mp_join = 1;
if (opsize == TCPOLEN_MPTCP_MPJ_SYN) {
mp_opt->backup = *ptr++ & MPTCPOPT_BACKUP;
mp_opt->join_id = *ptr++;
mp_opt->token = get_unaligned_be32(ptr);
ptr += 4;
mp_opt->nonce = get_unaligned_be32(ptr);
ptr += 4;
pr_debug("MP_JOIN bkup=%u, id=%u, token=%u, nonce=%u",
mp_opt->backup, mp_opt->join_id,
mp_opt->token, mp_opt->nonce);
} else if (opsize == TCPOLEN_MPTCP_MPJ_SYNACK) {
mp_opt->backup = *ptr++ & MPTCPOPT_BACKUP;
mp_opt->join_id = *ptr++;
mp_opt->thmac = get_unaligned_be64(ptr);
ptr += 8;
mp_opt->nonce = get_unaligned_be32(ptr);
ptr += 4;
pr_debug("MP_JOIN bkup=%u, id=%u, thmac=%llu, nonce=%u",
mp_opt->backup, mp_opt->join_id,
mp_opt->thmac, mp_opt->nonce);
} else if (opsize == TCPOLEN_MPTCP_MPJ_ACK) {
ptr += 2;
memcpy(mp_opt->hmac, ptr, MPTCPOPT_HMAC_LEN);
pr_debug("MP_JOIN hmac");
} else {
pr_warn("MP_JOIN bad option size");
mp_opt->mp_join = 0;
}
break;
case MPTCPOPT_DSS: case MPTCPOPT_DSS:
pr_debug("DSS"); pr_debug("DSS");
ptr++; ptr++;
...@@ -572,37 +604,80 @@ bool mptcp_synack_options(const struct request_sock *req, unsigned int *size, ...@@ -572,37 +604,80 @@ bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
pr_debug("subflow_req=%p, local_key=%llu", pr_debug("subflow_req=%p, local_key=%llu",
subflow_req, subflow_req->local_key); subflow_req, subflow_req->local_key);
return true; return true;
} else if (subflow_req->mp_join) {
opts->suboptions = OPTION_MPTCP_MPJ_SYNACK;
opts->backup = subflow_req->backup;
opts->join_id = subflow_req->local_id;
opts->thmac = subflow_req->thmac;
opts->nonce = subflow_req->local_nonce;
pr_debug("req=%p, bkup=%u, id=%u, thmac=%llu, nonce=%u",
subflow_req, opts->backup, opts->join_id,
opts->thmac, opts->nonce);
*size = TCPOLEN_MPTCP_MPJ_SYNACK;
return true;
} }
return false; return false;
} }
static bool check_fully_established(struct mptcp_subflow_context *subflow, static bool check_fully_established(struct mptcp_sock *msk, struct sock *sk,
struct mptcp_subflow_context *subflow,
struct sk_buff *skb, struct sk_buff *skb,
struct mptcp_options_received *mp_opt) struct mptcp_options_received *mp_opt)
{ {
/* here we can process OoO, in-window pkts, only in-sequence 4th ack /* here we can process OoO, in-window pkts, only in-sequence 4th ack
* are relevant * will make the subflow fully established
*/ */
if (likely(subflow->fully_established || if (likely(subflow->fully_established)) {
TCP_SKB_CB(skb)->seq != subflow->ssn_offset + 1)) /* on passive sockets, check for 3rd ack retransmission
return true; * note that msk is always set by subflow_syn_recv_sock()
* for mp_join subflows
*/
if (TCP_SKB_CB(skb)->seq == subflow->ssn_offset + 1 &&
TCP_SKB_CB(skb)->end_seq == TCP_SKB_CB(skb)->seq &&
subflow->mp_join && mp_opt->mp_join &&
READ_ONCE(msk->pm.server_side))
tcp_send_ack(sk);
goto fully_established;
}
/* we should process OoO packets before the first subflow is fully
* established, but not expected for MP_JOIN subflows
*/
if (TCP_SKB_CB(skb)->seq != subflow->ssn_offset + 1)
return subflow->mp_capable;
if (mp_opt->use_ack) if (mp_opt->use_ack) {
/* subflows are fully established as soon as we get any
* additional ack.
*/
subflow->fully_established = 1; subflow->fully_established = 1;
goto fully_established;
}
if (subflow->can_ack) WARN_ON_ONCE(subflow->can_ack);
return true;
/* If the first established packet does not contain MP_CAPABLE + data /* If the first established packet does not contain MP_CAPABLE + data
* then fallback to TCP * then fallback to TCP
*/ */
if (!mp_opt->mp_capable) { if (!mp_opt->mp_capable) {
subflow->mp_capable = 0; subflow->mp_capable = 0;
tcp_sk(mptcp_subflow_tcp_sock(subflow))->is_mptcp = 0; tcp_sk(sk)->is_mptcp = 0;
return false; return false;
} }
subflow->fully_established = 1;
subflow->remote_key = mp_opt->sndr_key; subflow->remote_key = mp_opt->sndr_key;
subflow->can_ack = 1; subflow->can_ack = 1;
fully_established:
if (likely(subflow->pm_notified))
return true;
subflow->pm_notified = 1;
if (subflow->mp_join)
mptcp_pm_subflow_established(msk, subflow);
else
mptcp_pm_fully_established(msk);
return true; return true;
} }
...@@ -641,7 +716,7 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb, ...@@ -641,7 +716,7 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
struct mptcp_ext *mpext; struct mptcp_ext *mpext;
mp_opt = &opt_rx->mptcp; mp_opt = &opt_rx->mptcp;
if (!check_fully_established(subflow, skb, mp_opt)) if (!check_fully_established(msk, sk, subflow, skb, mp_opt))
return; return;
if (mp_opt->add_addr && add_addr_hmac_valid(msk, mp_opt)) { if (mp_opt->add_addr && add_addr_hmac_valid(msk, mp_opt)) {
...@@ -700,8 +775,6 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb, ...@@ -700,8 +775,6 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
} }
mpext->data_fin = mp_opt->data_fin; mpext->data_fin = mp_opt->data_fin;
mptcp_pm_fully_established(msk);
} }
void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts) void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
...@@ -787,6 +860,16 @@ void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts) ...@@ -787,6 +860,16 @@ void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
0, opts->rm_id); 0, opts->rm_id);
} }
if (OPTION_MPTCP_MPJ_SYNACK & opts->suboptions) {
*ptr++ = mptcp_option(MPTCPOPT_MP_JOIN,
TCPOLEN_MPTCP_MPJ_SYNACK,
opts->backup, opts->join_id);
put_unaligned_be64(opts->thmac, ptr);
ptr += 2;
put_unaligned_be32(opts->nonce, ptr);
ptr += 1;
}
if (opts->ext_copy.use_ack || opts->ext_copy.use_map) { if (opts->ext_copy.use_ack || opts->ext_copy.use_map) {
struct mptcp_ext *mpext = &opts->ext_copy; struct mptcp_ext *mpext = &opts->ext_copy;
u8 len = TCPOLEN_MPTCP_DSS_BASE; u8 len = TCPOLEN_MPTCP_DSS_BASE;
......
...@@ -104,19 +104,6 @@ static struct socket *__mptcp_socket_create(struct mptcp_sock *msk, int state) ...@@ -104,19 +104,6 @@ static struct socket *__mptcp_socket_create(struct mptcp_sock *msk, int state)
return ssock; return ssock;
} }
static struct sock *mptcp_subflow_get(const struct mptcp_sock *msk)
{
struct mptcp_subflow_context *subflow;
sock_owned_by_me((const struct sock *)msk);
mptcp_for_each_subflow(msk, subflow) {
return mptcp_subflow_tcp_sock(subflow);
}
return NULL;
}
static void __mptcp_move_skb(struct mptcp_sock *msk, struct sock *ssk, static void __mptcp_move_skb(struct mptcp_sock *msk, struct sock *ssk,
struct sk_buff *skb, struct sk_buff *skb,
unsigned int offset, size_t copy_len) unsigned int offset, size_t copy_len)
...@@ -391,6 +378,43 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk, ...@@ -391,6 +378,43 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
return ret; return ret;
} }
static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)
{
struct mptcp_subflow_context *subflow;
struct sock *backup = NULL;
sock_owned_by_me((const struct sock *)msk);
mptcp_for_each_subflow(msk, subflow) {
struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
if (!sk_stream_memory_free(ssk)) {
struct socket *sock = ssk->sk_socket;
if (sock) {
clear_bit(MPTCP_SEND_SPACE, &msk->flags);
smp_mb__after_atomic();
/* enables sk->write_space() callbacks */
set_bit(SOCK_NOSPACE, &sock->flags);
}
return NULL;
}
if (subflow->backup) {
if (!backup)
backup = ssk;
continue;
}
return ssk;
}
return backup;
}
static void ssk_check_wmem(struct mptcp_sock *msk, struct sock *ssk) static void ssk_check_wmem(struct mptcp_sock *msk, struct sock *ssk)
{ {
struct socket *sock; struct socket *sock;
...@@ -438,10 +462,17 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) ...@@ -438,10 +462,17 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
return ret >= 0 ? ret + copied : (copied ? copied : ret); return ret >= 0 ? ret + copied : (copied ? copied : ret);
} }
ssk = mptcp_subflow_get(msk); ssk = mptcp_subflow_get_send(msk);
if (!ssk) { while (!sk_stream_memory_free(sk) || !ssk) {
release_sock(sk); ret = sk_stream_wait_memory(sk, &timeo);
return -ENOTCONN; if (ret)
goto out;
ssk = mptcp_subflow_get_send(msk);
if (list_empty(&msk->conn_list)) {
ret = -ENOTCONN;
goto out;
}
} }
pr_debug("conn_list->subflow=%p", ssk); pr_debug("conn_list->subflow=%p", ssk);
...@@ -1070,6 +1101,37 @@ static void mptcp_sock_graft(struct sock *sk, struct socket *parent) ...@@ -1070,6 +1101,37 @@ static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
write_unlock_bh(&sk->sk_callback_lock); write_unlock_bh(&sk->sk_callback_lock);
} }
bool mptcp_finish_join(struct sock *sk)
{
struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
struct mptcp_sock *msk = mptcp_sk(subflow->conn);
struct sock *parent = (void *)msk;
struct socket *parent_sock;
pr_debug("msk=%p, subflow=%p", msk, subflow);
/* mptcp socket already closing? */
if (inet_sk_state_load(parent) != TCP_ESTABLISHED)
return false;
if (!msk->pm.server_side)
return true;
/* passive connection, attach to msk socket */
parent_sock = READ_ONCE(parent->sk_socket);
if (parent_sock && !sk->sk_socket)
mptcp_sock_graft(sk, parent_sock);
return mptcp_pm_allow_new_subflow(msk);
}
bool mptcp_sk_is_subflow(const struct sock *sk)
{
struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
return subflow->mp_join == 1;
}
static bool mptcp_memory_free(const struct sock *sk, int wake) static bool mptcp_memory_free(const struct sock *sk, int wake)
{ {
struct mptcp_sock *msk = mptcp_sk(sk); struct mptcp_sock *msk = mptcp_sk(sk);
......
...@@ -17,6 +17,9 @@ ...@@ -17,6 +17,9 @@
#define OPTION_MPTCP_MPC_SYN BIT(0) #define OPTION_MPTCP_MPC_SYN BIT(0)
#define OPTION_MPTCP_MPC_SYNACK BIT(1) #define OPTION_MPTCP_MPC_SYNACK BIT(1)
#define OPTION_MPTCP_MPC_ACK BIT(2) #define OPTION_MPTCP_MPC_ACK BIT(2)
#define OPTION_MPTCP_MPJ_SYN BIT(3)
#define OPTION_MPTCP_MPJ_SYNACK BIT(4)
#define OPTION_MPTCP_MPJ_ACK BIT(5)
#define OPTION_MPTCP_ADD_ADDR BIT(6) #define OPTION_MPTCP_ADD_ADDR BIT(6)
#define OPTION_MPTCP_ADD_ADDR6 BIT(7) #define OPTION_MPTCP_ADD_ADDR6 BIT(7)
#define OPTION_MPTCP_RM_ADDR BIT(8) #define OPTION_MPTCP_RM_ADDR BIT(8)
...@@ -36,6 +39,9 @@ ...@@ -36,6 +39,9 @@
#define TCPOLEN_MPTCP_MPC_SYNACK 12 #define TCPOLEN_MPTCP_MPC_SYNACK 12
#define TCPOLEN_MPTCP_MPC_ACK 20 #define TCPOLEN_MPTCP_MPC_ACK 20
#define TCPOLEN_MPTCP_MPC_ACK_DATA 22 #define TCPOLEN_MPTCP_MPC_ACK_DATA 22
#define TCPOLEN_MPTCP_MPJ_SYN 12
#define TCPOLEN_MPTCP_MPJ_SYNACK 16
#define TCPOLEN_MPTCP_MPJ_ACK 24
#define TCPOLEN_MPTCP_DSS_BASE 4 #define TCPOLEN_MPTCP_DSS_BASE 4
#define TCPOLEN_MPTCP_DSS_ACK32 4 #define TCPOLEN_MPTCP_DSS_ACK32 4
#define TCPOLEN_MPTCP_DSS_ACK64 8 #define TCPOLEN_MPTCP_DSS_ACK64 8
...@@ -53,6 +59,9 @@ ...@@ -53,6 +59,9 @@
#define TCPOLEN_MPTCP_PORT_LEN 2 #define TCPOLEN_MPTCP_PORT_LEN 2
#define TCPOLEN_MPTCP_RM_ADDR_BASE 4 #define TCPOLEN_MPTCP_RM_ADDR_BASE 4
#define MPTCPOPT_BACKUP BIT(0)
#define MPTCPOPT_HMAC_LEN 20
/* MPTCP MP_CAPABLE flags */ /* MPTCP MP_CAPABLE flags */
#define MPTCP_VERSION_MASK (0x0F) #define MPTCP_VERSION_MASK (0x0F)
#define MPTCP_CAP_CHECKSUM_REQD BIT(7) #define MPTCP_CAP_CHECKSUM_REQD BIT(7)
...@@ -162,11 +171,15 @@ struct mptcp_subflow_request_sock { ...@@ -162,11 +171,15 @@ struct mptcp_subflow_request_sock {
backup : 1, backup : 1,
remote_key_valid : 1; remote_key_valid : 1;
u8 local_id; u8 local_id;
u8 remote_id;
u64 local_key; u64 local_key;
u64 remote_key; u64 remote_key;
u64 idsn; u64 idsn;
u32 token; u32 token;
u32 ssn_offset; u32 ssn_offset;
u64 thmac;
u32 local_nonce;
u32 remote_nonce;
}; };
static inline struct mptcp_subflow_request_sock * static inline struct mptcp_subflow_request_sock *
...@@ -190,15 +203,23 @@ struct mptcp_subflow_context { ...@@ -190,15 +203,23 @@ struct mptcp_subflow_context {
u32 map_data_len; u32 map_data_len;
u32 request_mptcp : 1, /* send MP_CAPABLE */ u32 request_mptcp : 1, /* send MP_CAPABLE */
mp_capable : 1, /* remote is MPTCP capable */ mp_capable : 1, /* remote is MPTCP capable */
mp_join : 1, /* remote is JOINing */
fully_established : 1, /* path validated */ fully_established : 1, /* path validated */
pm_notified : 1, /* PM hook called for established status */
conn_finished : 1, conn_finished : 1,
map_valid : 1, map_valid : 1,
mpc_map : 1, mpc_map : 1,
backup : 1,
data_avail : 1, data_avail : 1,
rx_eof : 1, rx_eof : 1,
data_fin_tx_enable : 1, data_fin_tx_enable : 1,
can_ack : 1; /* only after processing the remote a key */ can_ack : 1; /* only after processing the remote a key */
u64 data_fin_tx_seq; u64 data_fin_tx_seq;
u32 remote_nonce;
u64 thmac;
u32 local_nonce;
u8 local_id;
u8 remote_id;
struct sock *tcp_sock; /* tcp sk backpointer */ struct sock *tcp_sock; /* tcp sk backpointer */
struct sock *conn; /* parent mptcp_sock */ struct sock *conn; /* parent mptcp_sock */
...@@ -270,11 +291,13 @@ void mptcp_get_options(const struct sk_buff *skb, ...@@ -270,11 +291,13 @@ void mptcp_get_options(const struct sk_buff *skb,
void mptcp_finish_connect(struct sock *sk); void mptcp_finish_connect(struct sock *sk);
void mptcp_data_ready(struct sock *sk, struct sock *ssk); void mptcp_data_ready(struct sock *sk, struct sock *ssk);
bool mptcp_finish_join(struct sock *sk);
int mptcp_token_new_request(struct request_sock *req); int mptcp_token_new_request(struct request_sock *req);
void mptcp_token_destroy_request(u32 token); void mptcp_token_destroy_request(u32 token);
int mptcp_token_new_connect(struct sock *sk); int mptcp_token_new_connect(struct sock *sk);
int mptcp_token_new_accept(u32 token, struct sock *conn); int mptcp_token_new_accept(u32 token, struct sock *conn);
struct mptcp_sock *mptcp_token_get_sock(u32 token);
void mptcp_token_destroy(u32 token); void mptcp_token_destroy(u32 token);
void mptcp_crypto_key_sha(u64 key, u32 *token, u64 *idsn); void mptcp_crypto_key_sha(u64 key, u32 *token, u64 *idsn);
......
...@@ -9,6 +9,7 @@ ...@@ -9,6 +9,7 @@
#include <linux/kernel.h> #include <linux/kernel.h>
#include <linux/module.h> #include <linux/module.h>
#include <linux/netdevice.h> #include <linux/netdevice.h>
#include <crypto/algapi.h>
#include <net/sock.h> #include <net/sock.h>
#include <net/inet_common.h> #include <net/inet_common.h>
#include <net/inet_hashtables.h> #include <net/inet_hashtables.h>
...@@ -47,6 +48,52 @@ static void subflow_req_destructor(struct request_sock *req) ...@@ -47,6 +48,52 @@ static void subflow_req_destructor(struct request_sock *req)
tcp_request_sock_ops.destructor(req); tcp_request_sock_ops.destructor(req);
} }
static void subflow_generate_hmac(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
void *hmac)
{
u8 msg[8];
put_unaligned_be32(nonce1, &msg[0]);
put_unaligned_be32(nonce2, &msg[4]);
mptcp_crypto_hmac_sha(key1, key2, msg, 8, hmac);
}
/* validate received token and create truncated hmac and nonce for SYN-ACK */
static bool subflow_token_join_request(struct request_sock *req,
const struct sk_buff *skb)
{
struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);
u8 hmac[MPTCPOPT_HMAC_LEN];
struct mptcp_sock *msk;
int local_id;
msk = mptcp_token_get_sock(subflow_req->token);
if (!msk) {
pr_debug("subflow_req=%p, token=%u - not found\n",
subflow_req, subflow_req->token);
return false;
}
local_id = mptcp_pm_get_local_id(msk, (struct sock_common *)req);
if (local_id < 0) {
sock_put((struct sock *)msk);
return false;
}
subflow_req->local_id = local_id;
get_random_bytes(&subflow_req->local_nonce, sizeof(u32));
subflow_generate_hmac(msk->local_key, msk->remote_key,
subflow_req->local_nonce,
subflow_req->remote_nonce, hmac);
subflow_req->thmac = get_unaligned_be64(hmac);
sock_put((struct sock *)msk);
return true;
}
static void subflow_init_req(struct request_sock *req, static void subflow_init_req(struct request_sock *req,
const struct sock *sk_listener, const struct sock *sk_listener,
struct sk_buff *skb) struct sk_buff *skb)
...@@ -61,6 +108,7 @@ static void subflow_init_req(struct request_sock *req, ...@@ -61,6 +108,7 @@ static void subflow_init_req(struct request_sock *req,
mptcp_get_options(skb, &rx_opt); mptcp_get_options(skb, &rx_opt);
subflow_req->mp_capable = 0; subflow_req->mp_capable = 0;
subflow_req->mp_join = 0;
subflow_req->remote_key_valid = 0; subflow_req->remote_key_valid = 0;
#ifdef CONFIG_TCP_MD5SIG #ifdef CONFIG_TCP_MD5SIG
...@@ -71,6 +119,9 @@ static void subflow_init_req(struct request_sock *req, ...@@ -71,6 +119,9 @@ static void subflow_init_req(struct request_sock *req,
return; return;
#endif #endif
if (rx_opt.mptcp.mp_capable && rx_opt.mptcp.mp_join)
return;
if (rx_opt.mptcp.mp_capable && listener->request_mptcp) { if (rx_opt.mptcp.mp_capable && listener->request_mptcp) {
int err; int err;
...@@ -79,6 +130,18 @@ static void subflow_init_req(struct request_sock *req, ...@@ -79,6 +130,18 @@ static void subflow_init_req(struct request_sock *req,
subflow_req->mp_capable = 1; subflow_req->mp_capable = 1;
subflow_req->ssn_offset = TCP_SKB_CB(skb)->seq; subflow_req->ssn_offset = TCP_SKB_CB(skb)->seq;
} else if (rx_opt.mptcp.mp_join && listener->request_mptcp) {
subflow_req->mp_join = 1;
subflow_req->backup = rx_opt.mptcp.backup;
subflow_req->remote_id = rx_opt.mptcp.join_id;
subflow_req->token = rx_opt.mptcp.token;
subflow_req->remote_nonce = rx_opt.mptcp.nonce;
pr_debug("token=%u, remote_nonce=%u", subflow_req->token,
subflow_req->remote_nonce);
if (!subflow_token_join_request(req, skb)) {
subflow_req->mp_join = 0;
// @@ need to trigger RST
}
} }
} }
...@@ -178,6 +241,32 @@ static int subflow_v6_conn_request(struct sock *sk, struct sk_buff *skb) ...@@ -178,6 +241,32 @@ static int subflow_v6_conn_request(struct sock *sk, struct sk_buff *skb)
} }
#endif #endif
/* validate hmac received in third ACK */
static bool subflow_hmac_valid(const struct request_sock *req,
const struct tcp_options_received *rx_opt)
{
const struct mptcp_subflow_request_sock *subflow_req;
u8 hmac[MPTCPOPT_HMAC_LEN];
struct mptcp_sock *msk;
bool ret;
subflow_req = mptcp_subflow_rsk(req);
msk = mptcp_token_get_sock(subflow_req->token);
if (!msk)
return false;
subflow_generate_hmac(msk->remote_key, msk->local_key,
subflow_req->remote_nonce,
subflow_req->local_nonce, hmac);
ret = true;
if (crypto_memneq(hmac, rx_opt->mptcp.hmac, sizeof(hmac)))
ret = false;
sock_put((struct sock *)msk);
return ret;
}
static struct sock *subflow_syn_recv_sock(const struct sock *sk, static struct sock *subflow_syn_recv_sock(const struct sock *sk,
struct sk_buff *skb, struct sk_buff *skb,
struct request_sock *req, struct request_sock *req,
...@@ -188,6 +277,7 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, ...@@ -188,6 +277,7 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
struct mptcp_subflow_context *listener = mptcp_subflow_ctx(sk); struct mptcp_subflow_context *listener = mptcp_subflow_ctx(sk);
struct mptcp_subflow_request_sock *subflow_req; struct mptcp_subflow_request_sock *subflow_req;
struct tcp_options_received opt_rx; struct tcp_options_received opt_rx;
bool fallback_is_fatal = false;
struct sock *new_msk = NULL; struct sock *new_msk = NULL;
struct sock *child; struct sock *child;
...@@ -221,6 +311,13 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, ...@@ -221,6 +311,13 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
new_msk = mptcp_sk_clone(listener->conn, req); new_msk = mptcp_sk_clone(listener->conn, req);
if (!new_msk) if (!new_msk)
subflow_req->mp_capable = 0; subflow_req->mp_capable = 0;
} else if (subflow_req->mp_join) {
fallback_is_fatal = true;
opt_rx.mptcp.mp_join = 0;
mptcp_get_options(skb, &opt_rx);
if (!opt_rx.mptcp.mp_join ||
!subflow_hmac_valid(req, &opt_rx))
return NULL;
} }
create_child: create_child:
...@@ -230,11 +327,14 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, ...@@ -230,11 +327,14 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
if (child && *own_req) { if (child && *own_req) {
struct mptcp_subflow_context *ctx = mptcp_subflow_ctx(child); struct mptcp_subflow_context *ctx = mptcp_subflow_ctx(child);
/* we have null ctx on TCP fallback, not fatal on MPC /* we have null ctx on TCP fallback, which is fatal on
* handshake * MPJ handshake
*/ */
if (!ctx) if (!ctx) {
if (fallback_is_fatal)
goto close_child;
goto out; goto out;
}
if (ctx->mp_capable) { if (ctx->mp_capable) {
/* new mpc subflow takes ownership of the newly /* new mpc subflow takes ownership of the newly
...@@ -244,6 +344,16 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, ...@@ -244,6 +344,16 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
mptcp_pm_new_connection(mptcp_sk(new_msk), 1); mptcp_pm_new_connection(mptcp_sk(new_msk), 1);
ctx->conn = new_msk; ctx->conn = new_msk;
new_msk = NULL; new_msk = NULL;
} else if (ctx->mp_join) {
struct mptcp_sock *owner;
owner = mptcp_token_get_sock(ctx->token);
if (!owner)
goto close_child;
ctx->conn = (struct sock *)owner;
if (!mptcp_finish_join(child))
goto close_child;
} }
} }
...@@ -252,6 +362,12 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, ...@@ -252,6 +362,12 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
if (unlikely(new_msk)) if (unlikely(new_msk))
sock_put(new_msk); sock_put(new_msk);
return child; return child;
close_child:
tcp_send_active_reset(child, GFP_ATOMIC);
inet_csk_prepare_forced_close(child);
tcp_done(child);
return NULL;
} }
static struct inet_connection_sock_af_ops subflow_specific; static struct inet_connection_sock_af_ops subflow_specific;
...@@ -566,7 +682,7 @@ static void subflow_data_ready(struct sock *sk) ...@@ -566,7 +682,7 @@ static void subflow_data_ready(struct sock *sk)
struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
struct sock *parent = subflow->conn; struct sock *parent = subflow->conn;
if (!subflow->mp_capable) { if (!subflow->mp_capable && !subflow->mp_join) {
subflow->tcp_data_ready(sk); subflow->tcp_data_ready(sk);
parent->sk_data_ready(parent); parent->sk_data_ready(parent);
...@@ -785,7 +901,8 @@ static void subflow_ulp_clone(const struct request_sock *req, ...@@ -785,7 +901,8 @@ static void subflow_ulp_clone(const struct request_sock *req,
struct mptcp_subflow_context *old_ctx = mptcp_subflow_ctx(newsk); struct mptcp_subflow_context *old_ctx = mptcp_subflow_ctx(newsk);
struct mptcp_subflow_context *new_ctx; struct mptcp_subflow_context *new_ctx;
if (!tcp_rsk(req)->is_mptcp || !subflow_req->mp_capable) { if (!tcp_rsk(req)->is_mptcp ||
(!subflow_req->mp_capable && !subflow_req->mp_join)) {
subflow_ulp_fallback(newsk, old_ctx); subflow_ulp_fallback(newsk, old_ctx);
return; return;
} }
...@@ -796,9 +913,6 @@ static void subflow_ulp_clone(const struct request_sock *req, ...@@ -796,9 +913,6 @@ static void subflow_ulp_clone(const struct request_sock *req,
return; return;
} }
/* see comments in subflow_syn_recv_sock(), MPTCP connection is fully
* established only after we receive the remote key
*/
new_ctx->conn_finished = 1; new_ctx->conn_finished = 1;
new_ctx->icsk_af_ops = old_ctx->icsk_af_ops; new_ctx->icsk_af_ops = old_ctx->icsk_af_ops;
new_ctx->tcp_data_ready = old_ctx->tcp_data_ready; new_ctx->tcp_data_ready = old_ctx->tcp_data_ready;
...@@ -807,14 +921,26 @@ static void subflow_ulp_clone(const struct request_sock *req, ...@@ -807,14 +921,26 @@ static void subflow_ulp_clone(const struct request_sock *req,
new_ctx->rel_write_seq = 1; new_ctx->rel_write_seq = 1;
new_ctx->tcp_sock = newsk; new_ctx->tcp_sock = newsk;
new_ctx->mp_capable = 1; if (subflow_req->mp_capable) {
new_ctx->fully_established = subflow_req->remote_key_valid; /* see comments in subflow_syn_recv_sock(), MPTCP connection
new_ctx->can_ack = subflow_req->remote_key_valid; * is fully established only after we receive the remote key
new_ctx->remote_key = subflow_req->remote_key; */
new_ctx->local_key = subflow_req->local_key; new_ctx->mp_capable = 1;
new_ctx->token = subflow_req->token; new_ctx->fully_established = subflow_req->remote_key_valid;
new_ctx->ssn_offset = subflow_req->ssn_offset; new_ctx->can_ack = subflow_req->remote_key_valid;
new_ctx->idsn = subflow_req->idsn; new_ctx->remote_key = subflow_req->remote_key;
new_ctx->local_key = subflow_req->local_key;
new_ctx->token = subflow_req->token;
new_ctx->ssn_offset = subflow_req->ssn_offset;
new_ctx->idsn = subflow_req->idsn;
} else if (subflow_req->mp_join) {
new_ctx->mp_join = 1;
new_ctx->fully_established = 1;
new_ctx->backup = subflow_req->backup;
new_ctx->local_id = subflow_req->local_id;
new_ctx->token = subflow_req->token;
new_ctx->thmac = subflow_req->thmac;
}
} }
static struct tcp_ulp_ops subflow_ulp_ops __read_mostly = { static struct tcp_ulp_ops subflow_ulp_ops __read_mostly = {
......
...@@ -140,6 +140,33 @@ int mptcp_token_new_accept(u32 token, struct sock *conn) ...@@ -140,6 +140,33 @@ int mptcp_token_new_accept(u32 token, struct sock *conn)
return err; return err;
} }
/**
* mptcp_token_get_sock - retrieve mptcp connection sock using its token
* @token: token of the mptcp connection to retrieve
*
* This function returns the mptcp connection structure with the given token.
* A reference count on the mptcp socket returned is taken.
*
* returns NULL if no connection with the given token value exists.
*/
struct mptcp_sock *mptcp_token_get_sock(u32 token)
{
struct sock *conn;
spin_lock_bh(&token_tree_lock);
conn = radix_tree_lookup(&token_tree, token);
if (conn) {
/* token still reserved? */
if (conn == (struct sock *)&token_used)
conn = NULL;
else
sock_hold(conn);
}
spin_unlock_bh(&token_tree_lock);
return mptcp_sk(conn);
}
/** /**
* mptcp_token_destroy_request - remove mptcp connection/token * mptcp_token_destroy_request - remove mptcp connection/token
* @token - token of mptcp connection to remove * @token - token of mptcp connection to remove
......
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