From 36cbd3dcc10384f813ec0814255f576c84f2bcd4 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Wed, 5 Aug 2009 10:42:58 -0700 Subject: net: mark read-only arrays as const String literals are constant, and usually, we can also tag the array of pointers const too, moving it to the .rodata section. Signed-off-by: Jan Engelhardt Signed-off-by: David S. Miller --- include/net/sctp/constants.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/net/sctp') diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h index b05b055..8bc25f7 100644 --- a/include/net/sctp/constants.h +++ b/include/net/sctp/constants.h @@ -241,7 +241,9 @@ const char *sctp_tname(const sctp_subtype_t); /* timeouts */ const char *sctp_pname(const sctp_subtype_t); /* primitives */ /* This is a table of printable names of sctp_state_t's. */ -extern const char *sctp_state_tbl[], *sctp_evttype_tbl[], *sctp_status_tbl[]; +extern const char *const sctp_state_tbl[]; +extern const char *const sctp_evttype_tbl[]; +extern const char *const sctp_status_tbl[]; /* Maximum chunk length considering padding requirements. */ enum { SCTP_MAX_CHUNK_LEN = ((1<<16) - sizeof(__u32)) }; -- cgit v1.1 From b4e8c6a7e6caa9c80edcbfb78fc50962b792d5f0 Mon Sep 17 00:00:00 2001 From: Rami Rosen Date: Thu, 30 Jul 2009 09:38:43 +0300 Subject: sctp: remove unused union (sctp_cmsg_data_t) definition This patch removes an unused union definition (sctp_cmsg_data_t) from include/net/sctp/user.h. Signed-off-by: Rami Rosen Signed-off-by: Vlad Yasevich --- include/net/sctp/user.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include/net/sctp') diff --git a/include/net/sctp/user.h b/include/net/sctp/user.h index 1580c04..be2334a 100644 --- a/include/net/sctp/user.h +++ b/include/net/sctp/user.h @@ -210,12 +210,6 @@ enum sctp_sinfo_flags { }; -typedef union { - __u8 raw; - struct sctp_initmsg init; - struct sctp_sndrcvinfo sndrcv; -} sctp_cmsg_data_t; - /* These are cmsg_types. */ typedef enum sctp_cmsg_type { SCTP_INIT, /* 5.2.1 SCTP Initiation Structure */ -- cgit v1.1 From bec9640bb0d451813b1bb1f2cc13a5bfb17c3e48 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Thu, 30 Jul 2009 18:08:28 -0400 Subject: sctp: Disallow new connection on a closing socket If a socket has a lot of association that are in the process of of being closed/aborted, it is possible for a remote to establish new associations during the time period that the old ones are shutting down. If this was a result of a close() call, there will be no socket and will cause a memory leak. We'll prevent this by setting the socket state to CLOSING and disallow new associations when in this state. Signed-off-by: Vlad Yasevich --- include/net/sctp/constants.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net/sctp') diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h index 8bc25f7..af8c150 100644 --- a/include/net/sctp/constants.h +++ b/include/net/sctp/constants.h @@ -231,7 +231,7 @@ typedef enum { SCTP_SS_LISTENING = TCP_LISTEN, SCTP_SS_ESTABLISHING = TCP_SYN_SENT, SCTP_SS_ESTABLISHED = TCP_ESTABLISHED, - SCTP_SS_DISCONNECTING = TCP_CLOSING, + SCTP_SS_CLOSING = TCP_CLOSING, } sctp_sock_state_t; /* These functions map various type to printable names. */ -- cgit v1.1 From 9c5c62be2f794c7cee533d856f9f34c3cf21ff1b Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 10 Aug 2009 13:51:03 -0400 Subject: sctp: Send user messages to the lower layer as one Currenlty, sctp breaks up user messages into fragments and sends each fragment to the lower layer by itself. This means that for each fragment we go all the way down the stack and back up. This also discourages bundling of multiple fragments when they can fit into a sigle packet (ex: due to user setting a low fragmentation threashold). We introduce a new command SCTP_CMD_SND_MSG and hand the whole message down state machine. The state machine and the side-effect parser will cork the queue, add all chunks from the message to the queue, and then un-cork the queue thus causing the chunks to get transmitted. Signed-off-by: Vlad Yasevich --- include/net/sctp/command.h | 3 +++ include/net/sctp/structs.h | 1 + 2 files changed, 4 insertions(+) (limited to 'include/net/sctp') diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h index 3b96680..8be5135 100644 --- a/include/net/sctp/command.h +++ b/include/net/sctp/command.h @@ -106,6 +106,7 @@ typedef enum { SCTP_CMD_ASSOC_SHKEY, /* generate the association shared keys */ SCTP_CMD_T1_RETRAN, /* Mark for retransmission after T1 timeout */ SCTP_CMD_UPDATE_INITTAG, /* Update peer inittag */ + SCTP_CMD_SEND_MSG, /* Send the whole use message */ SCTP_CMD_LAST } sctp_verb_t; @@ -139,6 +140,7 @@ typedef union { struct sctp_ulpevent *ulpevent; struct sctp_packet *packet; sctp_sackhdr_t *sackh; + struct sctp_datamsg *msg; } sctp_arg_t; /* We are simulating ML type constructors here. @@ -188,6 +190,7 @@ SCTP_ARG_CONSTRUCTOR(PEER_INIT, sctp_init_chunk_t *, init) SCTP_ARG_CONSTRUCTOR(ULPEVENT, struct sctp_ulpevent *, ulpevent) SCTP_ARG_CONSTRUCTOR(PACKET, struct sctp_packet *, packet) SCTP_ARG_CONSTRUCTOR(SACKH, sctp_sackhdr_t *, sackh) +SCTP_ARG_CONSTRUCTOR(DATAMSG, struct sctp_datamsg *, msg) typedef struct { sctp_arg_t obj; diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index edfcacf..97024fa 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -643,6 +643,7 @@ struct sctp_datamsg { struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *, struct sctp_sndrcvinfo *, struct msghdr *, int len); +void sctp_datamsg_free(struct sctp_datamsg *); void sctp_datamsg_put(struct sctp_datamsg *); void sctp_chunk_fail(struct sctp_chunk *, int error); int sctp_chunk_abandoned(struct sctp_chunk *); -- cgit v1.1 From 4d3c46e6833208428d366630aa708f6876e61fc1 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Fri, 4 Sep 2009 18:20:59 -0400 Subject: sctp: drop a_rwnd to 0 when receive buffer overflows. SCTP has a problem that when small chunks are used, it is possible to exhaust the receiver buffer without fully closing receive window. This happens due to all overhead that we have account for with small messages. To fix this, when receive buffer is exceeded, we'll drop the window to 0 and save the 'drop' portion. When application starts reading data and freeing up recevie buffer space, we'll wait until we've reached the 'drop' window and then add back this 'drop' one mtu at a time. This worked well in testing and under stress produced rather even recovery. Signed-off-by: Vlad Yasevich --- include/net/sctp/structs.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/net/sctp') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 97024fa..b1bd268 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1739,6 +1739,12 @@ struct sctp_association { */ __u32 rwnd_over; + /* Keeps treack of rwnd pressure. This happens when we have + * a window, but not recevie buffer (i.e small packets). This one + * is releases slowly (1 PMTU at a time ). + */ + __u32 rwnd_press; + /* This is the sndbuf size in use for the association. * This corresponds to the sndbuf size for the association, * as specified in the sk->sndbuf. -- cgit v1.1 From cb95ea32a457871f72752164de8d94fa20f4703c Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Fri, 4 Sep 2009 18:20:59 -0400 Subject: sctp: Don't do NAGLE delay on large writes that were fragmented small SCTP will delay the last part of a large write due to NAGLE, if that part is smaller then MTU. Since we are doing large writes, we might as well send the last portion now instead of waiting untill the next large write happens. The small portion will be sent as is regardless, so it's better to not delay it. This is a result of much discussions with Wei Yongjun and Doug Graham . Many thanks go out to them. Signed-off-by: Vlad Yasevich --- include/net/sctp/structs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net/sctp') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index b1bd268..df4c632 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -628,7 +628,7 @@ struct sctp_datamsg { /* Chunks waiting to be submitted to lower layer. */ struct list_head chunks; /* Chunks that have been transmitted. */ - struct list_head track; + size_t msg_size; /* Reference counting. */ atomic_t refcnt; /* When is this message no longer interesting to the peer? */ -- cgit v1.1 From f68b2e05f326971cd76c65aa91a1a41771dd7485 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Fri, 4 Sep 2009 18:21:00 -0400 Subject: sctp: Fix SCTP_MAXSEG socket option to comply to spec. We had a bug that we never stored the user-defined value for MAXSEG when setting the value on an association. Thus future PMTU events ended up re-writing the frag point and increasing it past user limit. Additionally, when setting the option on the socket/endpoint, we effect all current associations, which is against spec. Now, we store the user 'maxseg' value along with the computed 'frag_point'. We inherit 'maxseg' from the socket at association creation and use it as an upper limit for 'frag_point' when its set. Signed-off-by: Vlad Yasevich --- include/net/sctp/sctp.h | 7 ++++--- include/net/sctp/structs.h | 1 + 2 files changed, 5 insertions(+), 3 deletions(-) (limited to 'include/net/sctp') diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index d16a304..8a6d529 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -486,15 +486,16 @@ static inline __s32 sctp_jitter(__u32 rto) } /* Break down data chunks at this point. */ -static inline int sctp_frag_point(const struct sctp_sock *sp, int pmtu) +static inline int sctp_frag_point(const struct sctp_association *asoc, int pmtu) { + struct sctp_sock *sp = sctp_sk(asoc->base.sk); int frag = pmtu; frag -= sp->pf->af->net_header_len; frag -= sizeof(struct sctphdr) + sizeof(struct sctp_data_chunk); - if (sp->user_frag) - frag = min_t(int, frag, sp->user_frag); + if (asoc->user_frag) + frag = min_t(int, frag, asoc->user_frag); frag = min_t(int, frag, SCTP_MAX_CHUNK_LEN); diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index df4c632..b106128 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1763,6 +1763,7 @@ struct sctp_association { /* The message size at which SCTP fragmentation will occur. */ __u32 frag_point; + __u32 user_frag; /* Counter used to count INIT errors. */ int init_err_counter; -- cgit v1.1 From a803c942303e6a4ef0ab6b80114529852cffa058 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Fri, 4 Sep 2009 18:21:01 -0400 Subject: sctp: Turn flags in 'sctp_packet' into bit fields This shrinks the size of sctp_packet a little. Signed-off-by: Vlad Yasevich --- include/net/sctp/structs.h | 22 ++++++---------------- 1 file changed, 6 insertions(+), 16 deletions(-) (limited to 'include/net/sctp') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index b106128..993cfff 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -812,22 +812,12 @@ struct sctp_packet { /* pointer to the auth chunk for this packet */ struct sctp_chunk *auth; - /* This packet contains a COOKIE-ECHO chunk. */ - __u8 has_cookie_echo; - - /* This packet contains a SACK chunk. */ - __u8 has_sack; - - /* This packet contains an AUTH chunk */ - __u8 has_auth; - - /* This packet contains at least 1 DATA chunk */ - __u8 has_data; - - /* SCTP cannot fragment this packet. So let ip fragment it. */ - __u8 ipfragok; - - __u8 malloced; + u8 has_cookie_echo:1, /* This packet contains a COOKIE-ECHO chunk. */ + has_sack:1, /* This packet contains a SACK chunk. */ + has_auth:1, /* This packet contains an AUTH chunk */ + has_data:1, /* This packet contains at least 1 DATA chunk */ + ipfragok:1, /* So let ip fragment this packet */ + malloced:1; /* Is it malloced? */ }; struct sctp_packet *sctp_packet_init(struct sctp_packet *, -- cgit v1.1 From 723884339f90a9c420783135168cc1045750eb5d Mon Sep 17 00:00:00 2001 From: Bhaskar Dutta Date: Thu, 3 Sep 2009 17:25:47 +0530 Subject: sctp: Sysctl configuration for IPv4 Address Scoping This patch introduces a new sysctl option to make IPv4 Address Scoping configurable . In networking environments where DNAT rules in iptables prerouting chains convert destination IP's to link-local/private IP addresses, SCTP connections fail to establish as the INIT chunk is dropped by the kernel due to address scope match failure. For example to support overlapping IP addresses (same IP address with different vlan id) a Layer-5 application listens on link local IP's, and there is a DNAT rule that maps the destination IP to a link local IP. Such applications never get the SCTP INIT if the address-scoping draft is strictly followed. This sysctl configuration allows SCTP to function in such unconventional networking environments. Sysctl options: 0 - Disable IPv4 address scoping draft altogether 1 - Enable IPv4 address scoping (default, current behavior) 2 - Enable address scoping but allow IPv4 private addresses in init/init-ack 3 - Enable address scoping but allow IPv4 link local address in init/init-ack Signed-off-by: Bhaskar Dutta Signed-off-by: Vlad Yasevich --- include/net/sctp/constants.h | 7 +++++++ include/net/sctp/structs.h | 10 ++++++++++ 2 files changed, 17 insertions(+) (limited to 'include/net/sctp') diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h index af8c150..58f714a 100644 --- a/include/net/sctp/constants.h +++ b/include/net/sctp/constants.h @@ -363,6 +363,13 @@ typedef enum { SCTP_SCOPE_UNUSABLE, /* IPv4 unusable addresses */ } sctp_scope_t; +typedef enum { + SCTP_SCOPE_POLICY_DISABLE, /* Disable IPv4 address scoping */ + SCTP_SCOPE_POLICY_ENABLE, /* Enable IPv4 address scoping */ + SCTP_SCOPE_POLICY_PRIVATE, /* Follow draft but allow IPv4 private addresses */ + SCTP_SCOPE_POLICY_LINK, /* Follow draft but allow IPv4 link local addresses */ +} sctp_scope_policy_t; + /* Based on IPv4 scoping , * SCTP IPv4 unusable addresses: 0.0.0.0/8, 224.0.0.0/4, 198.18.0.0/24, * 192.88.99.0/24. diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 993cfff..a48d80e 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -219,6 +219,15 @@ extern struct sctp_globals { /* Flag to idicate if SCTP-AUTH is enabled */ int auth_enable; + /* + * Policy to control SCTP IPv4 address scoping + * 0 - Disable IPv4 address scoping + * 1 - Enable IPv4 address scoping + * 2 - Selectively allow only IPv4 private addresses + * 3 - Selectively allow only IPv4 link local address + */ + int ipv4_scope_policy; + /* Flag to indicate whether computing and verifying checksum * is disabled. */ int checksum_disable; @@ -252,6 +261,7 @@ extern struct sctp_globals { #define sctp_port_hashtable (sctp_globals.port_hashtable) #define sctp_local_addr_list (sctp_globals.local_addr_list) #define sctp_local_addr_lock (sctp_globals.addr_list_lock) +#define sctp_scope_policy (sctp_globals.ipv4_scope_policy) #define sctp_addip_enable (sctp_globals.addip_enable) #define sctp_addip_noauth (sctp_globals.addip_noauth_enable) #define sctp_prsctp_enable (sctp_globals.prsctp_enable) -- cgit v1.1 From 9237ccbc0b22b5aa5396463ba0b14dc5efe5b98c Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Fri, 4 Sep 2009 14:33:19 +0800 Subject: sctp: turn flags in 'struct sctp_association' into bit fields This shrinks the size of struct sctp_association a little. Signed-off-by: Wei Yongjun Signed-off-by: Vlad Yasevich --- include/net/sctp/structs.h | 21 +++++++++------------ 1 file changed, 9 insertions(+), 12 deletions(-) (limited to 'include/net/sctp') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index a48d80e..42d00ce 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1568,13 +1568,13 @@ struct sctp_association { __u32 sack_cnt; /* These are capabilities which our peer advertised. */ - __u8 ecn_capable; /* Can peer do ECN? */ - __u8 ipv4_address; /* Peer understands IPv4 addresses? */ - __u8 ipv6_address; /* Peer understands IPv6 addresses? */ - __u8 hostname_address;/* Peer understands DNS addresses? */ - __u8 asconf_capable; /* Does peer support ADDIP? */ - __u8 prsctp_capable; /* Can peer do PR-SCTP? */ - __u8 auth_capable; /* Is peer doing SCTP-AUTH? */ + __u8 ecn_capable:1, /* Can peer do ECN? */ + ipv4_address:1, /* Peer understands IPv4 addresses? */ + ipv6_address:1, /* Peer understands IPv6 addresses? */ + hostname_address:1, /* Peer understands DNS addresses? */ + asconf_capable:1, /* Does peer support ADDIP? */ + prsctp_capable:1, /* Can peer do PR-SCTP? */ + auth_capable:1; /* Is peer doing SCTP-AUTH? */ __u32 adaptation_ind; /* Adaptation Code point. */ @@ -1913,11 +1913,8 @@ struct sctp_association { __u16 active_key_id; - /* Need to send an ECNE Chunk? */ - char need_ecne; - - /* Is it a temporary association? */ - char temp; + __u8 need_ecne:1, /* Need to send an ECNE Chunk? */ + temp:1; /* Is it a temporary association? */ }; -- cgit v1.1