diff options
Diffstat (limited to 'net/sctp/output.c')
| -rw-r--r-- | net/sctp/output.c | 830 |
1 files changed, 450 insertions, 380 deletions
diff --git a/net/sctp/output.c b/net/sctp/output.c index a46d1eb41762..23e96305cad7 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* SCTP kernel implementation * (C) Copyright IBM Corp. 2001, 2004 * Copyright (c) 1999-2000 Cisco, Inc. @@ -7,38 +8,15 @@ * * These functions handle output processing. * - * This SCTP implementation is free software; - * you can redistribute it and/or modify it under the terms of - * the GNU General Public License as published by - * the Free Software Foundation; either version 2, or (at your option) - * any later version. - * - * This SCTP implementation is distributed in the hope that it - * will be useful, but WITHOUT ANY WARRANTY; without even the implied - * ************************ - * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * See the GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with GNU CC; see the file COPYING. If not, write to - * the Free Software Foundation, 59 Temple Place - Suite 330, - * Boston, MA 02111-1307, USA. - * * Please send any bug reports or fixes you make to the * email address(es): - * lksctp developers <lksctp-developers@lists.sourceforge.net> - * - * Or submit a bug report through the following website: - * http://www.sf.net/projects/lksctp + * lksctp developers <linux-sctp@vger.kernel.org> * * Written or modified by: * La Monte H.P. Yarroll <piggy@acm.org> * Karl Knutson <karl@athena.chicago.il.us> * Jon Grimm <jgrimm@austin.ibm.com> * Sridhar Samudrala <sri@us.ibm.com> - * - * Any bugs reported given to us we will try to fix... any fixes shared will - * be incorporated into the next SCTP release. */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt @@ -64,19 +42,23 @@ #include <net/sctp/checksum.h> /* Forward declarations for private helpers. */ -static sctp_xmit_t __sctp_packet_append_chunk(struct sctp_packet *packet, - struct sctp_chunk *chunk); -static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet, - struct sctp_chunk *chunk); +static enum sctp_xmit __sctp_packet_append_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk); +static enum sctp_xmit sctp_packet_can_append_data(struct sctp_packet *packet, + struct sctp_chunk *chunk); static void sctp_packet_append_data(struct sctp_packet *packet, - struct sctp_chunk *chunk); -static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, - struct sctp_chunk *chunk, - u16 chunk_len); + struct sctp_chunk *chunk); +static enum sctp_xmit sctp_packet_will_fit(struct sctp_packet *packet, + struct sctp_chunk *chunk, + u16 chunk_len); static void sctp_packet_reset(struct sctp_packet *packet) { + /* sctp_packet_transmit() relies on this to reset size to the + * current overhead after sending packets. + */ packet->size = packet->overhead; + packet->has_cookie_echo = 0; packet->has_sack = 0; packet->has_data = 0; @@ -88,54 +70,91 @@ static void sctp_packet_reset(struct sctp_packet *packet) /* Config a packet. * This appears to be a followup set of initializations. */ -struct sctp_packet *sctp_packet_config(struct sctp_packet *packet, - __u32 vtag, int ecn_capable) +void sctp_packet_config(struct sctp_packet *packet, __u32 vtag, + int ecn_capable) { - struct sctp_chunk *chunk = NULL; + struct sctp_transport *tp = packet->transport; + struct sctp_association *asoc = tp->asoc; + struct sctp_sock *sp = NULL; + struct sock *sk; pr_debug("%s: packet:%p vtag:0x%x\n", __func__, packet, vtag); - packet->vtag = vtag; - if (ecn_capable && sctp_packet_empty(packet)) { - chunk = sctp_get_ecne_prepend(packet->transport->asoc); + /* do the following jobs only once for a flush schedule */ + if (!sctp_packet_empty(packet)) + return; + + /* set packet max_size with pathmtu, then calculate overhead */ + packet->max_size = tp->pathmtu; + + if (asoc) { + sk = asoc->base.sk; + sp = sctp_sk(sk); + } + packet->overhead = sctp_mtu_payload(sp, 0, 0); + packet->size = packet->overhead; + + if (!asoc) + return; + + /* update dst or transport pathmtu if in need */ + if (!sctp_transport_dst_check(tp)) { + sctp_transport_route(tp, NULL, sp); + if (asoc->param_flags & SPP_PMTUD_ENABLE) + sctp_assoc_sync_pmtu(asoc); + } else if (!sctp_transport_pl_enabled(tp) && + asoc->param_flags & SPP_PMTUD_ENABLE) { + if (!sctp_transport_pmtu_check(tp)) + sctp_assoc_sync_pmtu(asoc); + } + + if (asoc->pmtu_pending) { + if (asoc->param_flags & SPP_PMTUD_ENABLE) + sctp_assoc_sync_pmtu(asoc); + asoc->pmtu_pending = 0; + } + + /* If there a is a prepend chunk stick it on the list before + * any other chunks get appended. + */ + if (ecn_capable) { + struct sctp_chunk *chunk = sctp_get_ecne_prepend(asoc); - /* If there a is a prepend chunk stick it on the list before - * any other chunks get appended. - */ if (chunk) sctp_packet_append_chunk(packet, chunk); } - return packet; + if (!tp->dst) + return; + + /* set packet max_size with gso_max_size if gso is enabled*/ + rcu_read_lock(); + if (__sk_dst_get(sk) != tp->dst) { + dst_hold(tp->dst); + sk_setup_caps(sk, tp->dst); + } + packet->max_size = sk_can_gso(sk) ? min(READ_ONCE(tp->dst->dev->gso_max_size), + GSO_LEGACY_MAX_SIZE) + : asoc->pathmtu; + rcu_read_unlock(); } /* Initialize the packet structure. */ -struct sctp_packet *sctp_packet_init(struct sctp_packet *packet, - struct sctp_transport *transport, - __u16 sport, __u16 dport) +void sctp_packet_init(struct sctp_packet *packet, + struct sctp_transport *transport, + __u16 sport, __u16 dport) { - struct sctp_association *asoc = transport->asoc; - size_t overhead; - pr_debug("%s: packet:%p transport:%p\n", __func__, packet, transport); packet->transport = transport; packet->source_port = sport; packet->destination_port = dport; INIT_LIST_HEAD(&packet->chunk_list); - if (asoc) { - struct sctp_sock *sp = sctp_sk(asoc->base.sk); - overhead = sp->pf->af->net_header_len; - } else { - overhead = sizeof(struct ipv6hdr); - } - overhead += sizeof(struct sctphdr); - packet->overhead = overhead; + /* The overhead will be calculated by sctp_packet_config() */ + packet->overhead = 0; sctp_packet_reset(packet); packet->vtag = 0; - - return packet; } /* Free a packet. */ @@ -158,19 +177,21 @@ void sctp_packet_free(struct sctp_packet *packet) * as it can fit in the packet, but any more data that does not fit in this * packet can be sent only after receiving the COOKIE_ACK. */ -sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet, - struct sctp_chunk *chunk, - int one_packet) +enum sctp_xmit sctp_packet_transmit_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk, + int one_packet, gfp_t gfp) { - sctp_xmit_t retval; - int error = 0; + enum sctp_xmit retval; - pr_debug("%s: packet:%p chunk:%p\n", __func__, packet, chunk); + pr_debug("%s: packet:%p size:%zu chunk:%p size:%d\n", __func__, + packet, packet->size, chunk, chunk->skb ? chunk->skb->len : -1); switch ((retval = (sctp_packet_append_chunk(packet, chunk)))) { case SCTP_XMIT_PMTU_FULL: if (!packet->has_cookie_echo) { - error = sctp_packet_transmit(packet); + int error = 0; + + error = sctp_packet_transmit(packet, gfp); if (error < 0) chunk->skb->sk->sk_err = -error; @@ -185,20 +206,44 @@ sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet, case SCTP_XMIT_RWND_FULL: case SCTP_XMIT_OK: - case SCTP_XMIT_NAGLE_DELAY: + case SCTP_XMIT_DELAY: break; } return retval; } +/* Try to bundle a pad chunk into a packet with a heartbeat chunk for PLPMTUTD probe */ +static enum sctp_xmit sctp_packet_bundle_pad(struct sctp_packet *pkt, struct sctp_chunk *chunk) +{ + struct sctp_transport *t = pkt->transport; + struct sctp_chunk *pad; + int overhead = 0; + + if (!chunk->pmtu_probe) + return SCTP_XMIT_OK; + + /* calculate the Padding Data size for the pad chunk */ + overhead += sizeof(struct sctphdr) + sizeof(struct sctp_chunkhdr); + overhead += sizeof(struct sctp_sender_hb_info) + sizeof(struct sctp_pad_chunk); + pad = sctp_make_pad(t->asoc, t->pl.probe_size - overhead); + if (!pad) + return SCTP_XMIT_DELAY; + + list_add_tail(&pad->list, &pkt->chunk_list); + pkt->size += SCTP_PAD4(ntohs(pad->chunk_hdr->length)); + chunk->transport = t; + + return SCTP_XMIT_OK; +} + /* Try to bundle an auth chunk into the packet. */ -static sctp_xmit_t sctp_packet_bundle_auth(struct sctp_packet *pkt, - struct sctp_chunk *chunk) +static enum sctp_xmit sctp_packet_bundle_auth(struct sctp_packet *pkt, + struct sctp_chunk *chunk) { struct sctp_association *asoc = pkt->transport->asoc; + enum sctp_xmit retval = SCTP_XMIT_OK; struct sctp_chunk *auth; - sctp_xmit_t retval = SCTP_XMIT_OK; /* if we don't have an association, we can't do authentication */ if (!asoc) @@ -216,10 +261,13 @@ static sctp_xmit_t sctp_packet_bundle_auth(struct sctp_packet *pkt, if (!chunk->auth) return retval; - auth = sctp_make_auth(asoc); + auth = sctp_make_auth(asoc, chunk->shkey->key_id); if (!auth) return retval; + auth->shkey = chunk->shkey; + sctp_auth_shkey_hold(auth->shkey); + retval = __sctp_packet_append_chunk(pkt, auth); if (retval != SCTP_XMIT_OK) @@ -229,10 +277,10 @@ static sctp_xmit_t sctp_packet_bundle_auth(struct sctp_packet *pkt, } /* Try to bundle a SACK with the packet. */ -static sctp_xmit_t sctp_packet_bundle_sack(struct sctp_packet *pkt, - struct sctp_chunk *chunk) +static enum sctp_xmit sctp_packet_bundle_sack(struct sctp_packet *pkt, + struct sctp_chunk *chunk) { - sctp_xmit_t retval = SCTP_XMIT_OK; + enum sctp_xmit retval = SCTP_XMIT_OK; /* If sending DATA and haven't aleady bundled a SACK, try to * bundle one in to the packet. @@ -260,8 +308,11 @@ static sctp_xmit_t sctp_packet_bundle_sack(struct sctp_packet *pkt, sctp_chunk_free(sack); goto out; } + SCTP_INC_STATS(asoc->base.net, + SCTP_MIB_OUTCTRLCHUNKS); + asoc->stats.octrlchunks++; asoc->peer.sack_needed = 0; - if (del_timer(timer)) + if (timer_delete(timer)) sctp_association_put(asoc); } } @@ -274,11 +325,11 @@ out: /* Append a chunk to the offered packet reporting back any inability to do * so. */ -static sctp_xmit_t __sctp_packet_append_chunk(struct sctp_packet *packet, - struct sctp_chunk *chunk) +static enum sctp_xmit __sctp_packet_append_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk) { - sctp_xmit_t retval = SCTP_XMIT_OK; - __u16 chunk_len = WORD_ROUND(ntohs(chunk->chunk_hdr->length)); + __u16 chunk_len = SCTP_PAD4(ntohs(chunk->chunk_hdr->length)); + enum sctp_xmit retval = SCTP_XMIT_OK; /* Check to see if this chunk will fit into the packet */ retval = sctp_packet_will_fit(packet, chunk, chunk_len); @@ -287,7 +338,8 @@ static sctp_xmit_t __sctp_packet_append_chunk(struct sctp_packet *packet, /* We believe that this chunk is OK to add to the packet */ switch (chunk->chunk_hdr->type) { - case SCTP_CID_DATA: + case SCTP_CID_DATA: + case SCTP_CID_I_DATA: /* Account for the data being in the packet */ sctp_packet_append_data(packet, chunk); /* Disallow SACK bundling after DATA. */ @@ -298,18 +350,20 @@ static sctp_xmit_t __sctp_packet_append_chunk(struct sctp_packet *packet, packet->has_data = 1; /* timestamp the chunk for rtx purposes */ chunk->sent_at = jiffies; + /* Mainly used for prsctp RTX policy */ + chunk->sent_count++; break; - case SCTP_CID_COOKIE_ECHO: + case SCTP_CID_COOKIE_ECHO: packet->has_cookie_echo = 1; break; - case SCTP_CID_SACK: + case SCTP_CID_SACK: packet->has_sack = 1; if (chunk->asoc) chunk->asoc->stats.osacks++; break; - case SCTP_CID_AUTH: + case SCTP_CID_AUTH: packet->has_auth = 1; packet->auth = chunk; break; @@ -326,10 +380,10 @@ finish: /* Append a chunk to the offered packet reporting back any inability to do * so. */ -sctp_xmit_t sctp_packet_append_chunk(struct sctp_packet *packet, - struct sctp_chunk *chunk) +enum sctp_xmit sctp_packet_append_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk) { - sctp_xmit_t retval = SCTP_XMIT_OK; + enum sctp_xmit retval = SCTP_XMIT_OK; pr_debug("%s: packet:%p chunk:%p\n", __func__, packet, chunk); @@ -354,287 +408,260 @@ sctp_xmit_t sctp_packet_append_chunk(struct sctp_packet *packet, goto finish; retval = __sctp_packet_append_chunk(packet, chunk); + if (retval != SCTP_XMIT_OK) + goto finish; + + retval = sctp_packet_bundle_pad(packet, chunk); finish: return retval; } -static void sctp_packet_release_owner(struct sk_buff *skb) +static void sctp_packet_gso_append(struct sk_buff *head, struct sk_buff *skb) { - sk_free(skb->sk); -} + if (SCTP_OUTPUT_CB(head)->last == head) + skb_shinfo(head)->frag_list = skb; + else + SCTP_OUTPUT_CB(head)->last->next = skb; + SCTP_OUTPUT_CB(head)->last = skb; -static void sctp_packet_set_owner_w(struct sk_buff *skb, struct sock *sk) -{ - skb_orphan(skb); - skb->sk = sk; - skb->destructor = sctp_packet_release_owner; - - /* - * The data chunks have already been accounted for in sctp_sendmsg(), - * therefore only reserve a single byte to keep socket around until - * the packet has been transmitted. - */ - atomic_inc(&sk->sk_wmem_alloc); + head->truesize += skb->truesize; + head->data_len += skb->len; + head->len += skb->len; + refcount_add(skb->truesize, &head->sk->sk_wmem_alloc); + + __skb_header_release(skb); } -/* All packets are sent to the network through this function from - * sctp_outq_tail(). - * - * The return value is a normal kernel error return value. - */ -int sctp_packet_transmit(struct sctp_packet *packet) +static int sctp_packet_pack(struct sctp_packet *packet, + struct sk_buff *head, int gso, gfp_t gfp) { struct sctp_transport *tp = packet->transport; - struct sctp_association *asoc = tp->asoc; - struct sctphdr *sh; - struct sk_buff *nskb; + struct sctp_auth_chunk *auth = NULL; struct sctp_chunk *chunk, *tmp; - struct sock *sk; - int err = 0; - int padding; /* How much padding do we need? */ - __u8 has_data = 0; - struct dst_entry *dst = tp->dst; - unsigned char *auth = NULL; /* pointer to auth in skb data */ - __u32 cksum_buf_len = sizeof(struct sctphdr); + int pkt_count = 0, pkt_size; + struct sock *sk = head->sk; + struct sk_buff *nskb; + int auth_len = 0; - pr_debug("%s: packet:%p\n", __func__, packet); + if (gso) { + skb_shinfo(head)->gso_type = sk->sk_gso_type; + SCTP_OUTPUT_CB(head)->last = head; + } else { + nskb = head; + pkt_size = packet->size; + goto merge; + } - /* Do NOT generate a chunkless packet. */ - if (list_empty(&packet->chunk_list)) - return err; + do { + /* calculate the pkt_size and alloc nskb */ + pkt_size = packet->overhead; + list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, + list) { + int padded = SCTP_PAD4(chunk->skb->len); + + if (chunk == packet->auth) + auth_len = padded; + else if (auth_len + padded + packet->overhead > + tp->pathmtu) + return 0; + else if (pkt_size + padded > tp->pathmtu) + break; + pkt_size += padded; + } + nskb = alloc_skb(pkt_size + MAX_HEADER, gfp); + if (!nskb) + return 0; + skb_reserve(nskb, packet->overhead + MAX_HEADER); + +merge: + /* merge chunks into nskb and append nskb into head list */ + pkt_size -= packet->overhead; + list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) { + int padding; + + list_del_init(&chunk->list); + if (sctp_chunk_is_data(chunk)) { + if (!sctp_chunk_retransmitted(chunk) && + !tp->rto_pending) { + chunk->rtt_in_progress = 1; + tp->rto_pending = 1; + } + } - /* Set up convenience variables... */ - chunk = list_entry(packet->chunk_list.next, struct sctp_chunk, list); - sk = chunk->skb->sk; + padding = SCTP_PAD4(chunk->skb->len) - chunk->skb->len; + if (padding) + skb_put_zero(chunk->skb, padding); - /* Allocate the new skb. */ - nskb = alloc_skb(packet->size + LL_MAX_HEADER, GFP_ATOMIC); - if (!nskb) - goto nomem; + if (chunk == packet->auth) + auth = (struct sctp_auth_chunk *) + skb_tail_pointer(nskb); - /* Make sure the outbound skb has enough header room reserved. */ - skb_reserve(nskb, packet->overhead + LL_MAX_HEADER); + skb_put_data(nskb, chunk->skb->data, chunk->skb->len); - /* Set the owning socket so that we know where to get the - * destination IP address. - */ - sctp_packet_set_owner_w(nskb, sk); + pr_debug("*** Chunk:%p[%s] %s 0x%x, length:%d, chunk->skb->len:%d, rtt_in_progress:%d\n", + chunk, + sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)), + chunk->has_tsn ? "TSN" : "No TSN", + chunk->has_tsn ? ntohl(chunk->subh.data_hdr->tsn) : 0, + ntohs(chunk->chunk_hdr->length), chunk->skb->len, + chunk->rtt_in_progress); - if (!sctp_transport_dst_check(tp)) { - sctp_transport_route(tp, NULL, sctp_sk(sk)); - if (asoc && (asoc->param_flags & SPP_PMTUD_ENABLE)) { - sctp_assoc_sync_pmtu(sk, asoc); - } - } - dst = dst_clone(tp->dst); - skb_dst_set(nskb, dst); - if (!dst) - goto no_route; - - /* Build the SCTP header. */ - sh = (struct sctphdr *)skb_push(nskb, sizeof(struct sctphdr)); - skb_reset_transport_header(nskb); - sh->source = htons(packet->source_port); - sh->dest = htons(packet->destination_port); + pkt_size -= SCTP_PAD4(chunk->skb->len); - /* From 6.8 Adler-32 Checksum Calculation: - * After the packet is constructed (containing the SCTP common - * header and one or more control or DATA chunks), the - * transmitter shall: - * - * 1) Fill in the proper Verification Tag in the SCTP common - * header and initialize the checksum field to 0's. - */ - sh->vtag = htonl(packet->vtag); - sh->checksum = 0; + if (!sctp_chunk_is_data(chunk) && chunk != packet->auth) + sctp_chunk_free(chunk); - /** - * 6.10 Bundling - * - * An endpoint bundles chunks by simply including multiple - * chunks in one outbound SCTP packet. ... - */ + if (!pkt_size) + break; + } - /** - * 3.2 Chunk Field Descriptions - * - * The total length of a chunk (including Type, Length and - * Value fields) MUST be a multiple of 4 bytes. If the length - * of the chunk is not a multiple of 4 bytes, the sender MUST - * pad the chunk with all zero bytes and this padding is not - * included in the chunk length field. The sender should - * never pad with more than 3 bytes. - * - * [This whole comment explains WORD_ROUND() below.] - */ + if (auth) { + sctp_auth_calculate_hmac(tp->asoc, nskb, auth, + packet->auth->shkey, gfp); + /* free auth if no more chunks, or add it back */ + if (list_empty(&packet->chunk_list)) + sctp_chunk_free(packet->auth); + else + list_add(&packet->auth->list, + &packet->chunk_list); + } - pr_debug("***sctp_transmit_packet***\n"); + if (gso) + sctp_packet_gso_append(head, nskb); - list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) { - list_del_init(&chunk->list); - if (sctp_chunk_is_data(chunk)) { - /* 6.3.1 C4) When data is in flight and when allowed - * by rule C5, a new RTT measurement MUST be made each - * round trip. Furthermore, new RTT measurements - * SHOULD be made no more than once per round-trip - * for a given destination transport address. - */ + pkt_count++; + } while (!list_empty(&packet->chunk_list)); - if (!tp->rto_pending) { - chunk->rtt_in_progress = 1; - tp->rto_pending = 1; - } - has_data = 1; - } + if (gso) { + memset(head->cb, 0, max(sizeof(struct inet_skb_parm), + sizeof(struct inet6_skb_parm))); + skb_shinfo(head)->gso_segs = pkt_count; + skb_shinfo(head)->gso_size = GSO_BY_FRAGS; + goto chksum; + } - padding = WORD_ROUND(chunk->skb->len) - chunk->skb->len; - if (padding) - memset(skb_put(chunk->skb, padding), 0, padding); + if (sctp_checksum_disable) + return 1; - /* if this is the auth chunk that we are adding, - * store pointer where it will be added and put - * the auth into the packet. - */ - if (chunk == packet->auth) - auth = skb_tail_pointer(nskb); - - cksum_buf_len += chunk->skb->len; - memcpy(skb_put(nskb, chunk->skb->len), - chunk->skb->data, chunk->skb->len); - - pr_debug("*** Chunk:%p[%s] %s 0x%x, length:%d, chunk->skb->len:%d, " - "rtt_in_progress:%d\n", chunk, - sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)), - chunk->has_tsn ? "TSN" : "No TSN", - chunk->has_tsn ? ntohl(chunk->subh.data_hdr->tsn) : 0, - ntohs(chunk->chunk_hdr->length), chunk->skb->len, - chunk->rtt_in_progress); - - /* - * If this is a control chunk, this is our last - * reference. Free data chunks after they've been - * acknowledged or have failed. - */ - if (!sctp_chunk_is_data(chunk)) - sctp_chunk_free(chunk); + if (!(tp->dst->dev->features & NETIF_F_SCTP_CRC) || + dst_xfrm(tp->dst) || packet->ipfragok || tp->encap_port) { + struct sctphdr *sh = + (struct sctphdr *)skb_transport_header(head); + + sh->checksum = sctp_compute_cksum(head, 0); + } else { +chksum: + head->ip_summed = CHECKSUM_PARTIAL; + head->csum_not_inet = 1; + head->csum_start = skb_transport_header(head) - head->head; + head->csum_offset = offsetof(struct sctphdr, checksum); } - /* SCTP-AUTH, Section 6.2 - * The sender MUST calculate the MAC as described in RFC2104 [2] - * using the hash function H as described by the MAC Identifier and - * the shared association key K based on the endpoint pair shared key - * described by the shared key identifier. The 'data' used for the - * computation of the AUTH-chunk is given by the AUTH chunk with its - * HMAC field set to zero (as shown in Figure 6) followed by all - * chunks that are placed after the AUTH chunk in the SCTP packet. - */ - if (auth) - sctp_auth_calculate_hmac(asoc, nskb, - (struct sctp_auth_chunk *)auth, - GFP_ATOMIC); - - /* 2) Calculate the Adler-32 checksum of the whole packet, - * including the SCTP common header and all the - * chunks. - * - * Note: Adler-32 is no longer applicable, as has been replaced - * by CRC32-C as described in <draft-ietf-tsvwg-sctpcsum-02.txt>. - */ - if (!sctp_checksum_disable) { - if (!(dst->dev->features & NETIF_F_SCTP_CSUM)) { - __u32 crc32 = sctp_start_cksum((__u8 *)sh, cksum_buf_len); + return pkt_count; +} - /* 3) Put the resultant value into the checksum field in the - * common header, and leave the rest of the bits unchanged. - */ - sh->checksum = sctp_end_cksum(crc32); +/* All packets are sent to the network through this function from + * sctp_outq_tail(). + * + * The return value is always 0 for now. + */ +int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) +{ + struct sctp_transport *tp = packet->transport; + struct sctp_association *asoc = tp->asoc; + struct sctp_chunk *chunk, *tmp; + int pkt_count, gso = 0; + struct sk_buff *head; + struct sctphdr *sh; + struct sock *sk; + + pr_debug("%s: packet:%p\n", __func__, packet); + if (list_empty(&packet->chunk_list)) + return 0; + chunk = list_entry(packet->chunk_list.next, struct sctp_chunk, list); + sk = chunk->skb->sk; + + if (packet->size > tp->pathmtu && !packet->ipfragok && !chunk->pmtu_probe) { + if (tp->pl.state == SCTP_PL_ERROR) { /* do IP fragmentation if in Error state */ + packet->ipfragok = 1; } else { - /* no need to seed pseudo checksum for SCTP */ - nskb->ip_summed = CHECKSUM_PARTIAL; - nskb->csum_start = (skb_transport_header(nskb) - - nskb->head); - nskb->csum_offset = offsetof(struct sctphdr, checksum); + if (!sk_can_gso(sk)) { /* check gso */ + pr_err_once("Trying to GSO but underlying device doesn't support it."); + goto out; + } + gso = 1; } } - /* IP layer ECN support - * From RFC 2481 - * "The ECN-Capable Transport (ECT) bit would be set by the - * data sender to indicate that the end-points of the - * transport protocol are ECN-capable." - * - * Now setting the ECT bit all the time, as it should not cause - * any problems protocol-wise even if our peer ignores it. - * - * Note: The works for IPv6 layer checks this bit too later - * in transmission. See IP6_ECN_flow_xmit(). - */ - (*tp->af_specific->ecn_capable)(nskb->sk); + /* alloc head skb */ + head = alloc_skb((gso ? packet->overhead : packet->size) + + MAX_HEADER, gfp); + if (!head) + goto out; + skb_reserve(head, packet->overhead + MAX_HEADER); + skb_set_owner_w(head, sk); + + /* set sctp header */ + sh = skb_push(head, sizeof(struct sctphdr)); + skb_reset_transport_header(head); + sh->source = htons(packet->source_port); + sh->dest = htons(packet->destination_port); + sh->vtag = htonl(packet->vtag); + sh->checksum = 0; - /* Set up the IP options. */ - /* BUG: not implemented - * For v4 this all lives somewhere in sk->sk_opt... - */ + /* drop packet if no dst */ + if (!tp->dst) { + IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTNOROUTES); + kfree_skb(head); + goto out; + } - /* Dump that on IP! */ + /* pack up chunks */ + pkt_count = sctp_packet_pack(packet, head, gso, gfp); + if (!pkt_count) { + kfree_skb(head); + goto out; + } + pr_debug("***sctp_transmit_packet*** skb->len:%d\n", head->len); + + /* start autoclose timer */ + if (packet->has_data && sctp_state(asoc, ESTABLISHED) && + asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) { + struct timer_list *timer = + &asoc->timers[SCTP_EVENT_TIMEOUT_AUTOCLOSE]; + unsigned long timeout = + asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]; + + if (!mod_timer(timer, jiffies + timeout)) + sctp_association_hold(asoc); + } + + /* sctp xmit */ + tp->af_specific->ecn_capable(sk); if (asoc) { - asoc->stats.opackets++; + asoc->stats.opackets += pkt_count; if (asoc->peer.last_sent_to != tp) - /* Considering the multiple CPU scenario, this is a - * "correcter" place for last_sent_to. --xguo - */ asoc->peer.last_sent_to = tp; } - - if (has_data) { - struct timer_list *timer; - unsigned long timeout; - - /* Restart the AUTOCLOSE timer when sending data. */ - if (sctp_state(asoc, ESTABLISHED) && asoc->autoclose) { - timer = &asoc->timers[SCTP_EVENT_TIMEOUT_AUTOCLOSE]; - timeout = asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]; - - if (!mod_timer(timer, jiffies + timeout)) - sctp_association_hold(asoc); - } - } - - pr_debug("***sctp_transmit_packet*** skb->len:%d\n", nskb->len); - - nskb->local_df = packet->ipfragok; - (*tp->af_specific->sctp_xmit)(nskb, tp); - -out: - sctp_packet_reset(packet); - return err; -no_route: - kfree_skb(nskb); - IP_INC_STATS_BH(sock_net(asoc->base.sk), IPSTATS_MIB_OUTNOROUTES); - - /* FIXME: Returning the 'err' will effect all the associations - * associated with a socket, although only one of the paths of the - * association is unreachable. - * The real failure of a transport or association can be passed on - * to the user via notifications. So setting this error may not be - * required. - */ - /* err = -EHOSTUNREACH; */ -err: - /* Control chunks are unreliable so just drop them. DATA chunks - * will get resent or dropped later. + head->ignore_df = packet->ipfragok; + if (tp->dst_pending_confirm) + skb_set_dst_pending_confirm(head, 1); + /* neighbour should be confirmed on successful transmission or + * positive error */ + if (tp->af_specific->sctp_xmit(head, tp) >= 0 && + tp->dst_pending_confirm) + tp->dst_pending_confirm = 0; +out: list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) { list_del_init(&chunk->list); if (!sctp_chunk_is_data(chunk)) sctp_chunk_free(chunk); } - goto out; -nomem: - err = -ENOMEM; - goto err; + sctp_packet_reset(packet); + return 0; } /******************************************************************** @@ -642,10 +669,9 @@ nomem: ********************************************************************/ /* This private function check to see if a chunk can be added */ -static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet, - struct sctp_chunk *chunk) +static enum sctp_xmit sctp_packet_can_append_data(struct sctp_packet *packet, + struct sctp_chunk *chunk) { - sctp_xmit_t retval = SCTP_XMIT_OK; size_t datasize, rwnd, inflight, flight_size; struct sctp_transport *transport = packet->transport; struct sctp_association *asoc = transport->asoc; @@ -670,15 +696,11 @@ static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet, datasize = sctp_data_size(chunk); - if (datasize > rwnd) { - if (inflight > 0) { - /* We have (at least) one data chunk in flight, - * so we can't fall back to rule 6.1 B). - */ - retval = SCTP_XMIT_RWND_FULL; - goto finish; - } - } + if (datasize > rwnd && inflight > 0) + /* We have (at least) one data chunk in flight, + * so we can't fall back to rule 6.1 B). + */ + return SCTP_XMIT_RWND_FULL; /* RFC 2960 6.1 Transmission of DATA Chunks * @@ -692,36 +714,42 @@ static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet, * When a Fast Retransmit is being performed the sender SHOULD * ignore the value of cwnd and SHOULD NOT delay retransmission. */ - if (chunk->fast_retransmit != SCTP_NEED_FRTX) - if (flight_size >= transport->cwnd) { - retval = SCTP_XMIT_RWND_FULL; - goto finish; - } + if (chunk->fast_retransmit != SCTP_NEED_FRTX && + flight_size >= transport->cwnd) + return SCTP_XMIT_RWND_FULL; /* Nagle's algorithm to solve small-packet problem: * Inhibit the sending of new chunks when new outgoing data arrives * if any previously transmitted data on the connection remains * unacknowledged. */ - if (!sctp_sk(asoc->base.sk)->nodelay && sctp_packet_empty(packet) && - inflight && sctp_state(asoc, ESTABLISHED)) { - unsigned int max = transport->pathmtu - packet->overhead; - unsigned int len = chunk->skb->len + q->out_qlen; - - /* Check whether this chunk and all the rest of pending - * data will fit or delay in hopes of bundling a full - * sized packet. - * Don't delay large message writes that may have been - * fragmeneted into small peices. - */ - if ((len < max) && chunk->msg->can_delay) { - retval = SCTP_XMIT_NAGLE_DELAY; - goto finish; - } - } -finish: - return retval; + if ((sctp_sk(asoc->base.sk)->nodelay || inflight == 0) && + !asoc->force_delay) + /* Nothing unacked */ + return SCTP_XMIT_OK; + + if (!sctp_packet_empty(packet)) + /* Append to packet */ + return SCTP_XMIT_OK; + + if (!sctp_state(asoc, ESTABLISHED)) + return SCTP_XMIT_OK; + + /* Check whether this chunk and all the rest of pending data will fit + * or delay in hopes of bundling a full sized packet. + */ + if (chunk->skb->len + q->out_qlen > transport->pathmtu - + packet->overhead - sctp_datachk_len(&chunk->asoc->stream) - 4) + /* Enough data queued to fill a packet */ + return SCTP_XMIT_OK; + + /* Don't delay large message writes that may have been fragmented */ + if (!chunk->msg->can_delay) + return SCTP_XMIT_OK; + + /* Defer until all data acked or packet full */ + return SCTP_XMIT_DELAY; } /* This private function does management things when adding DATA chunk */ @@ -746,50 +774,92 @@ static void sctp_packet_append_data(struct sctp_packet *packet, rwnd = 0; asoc->peer.rwnd = rwnd; - /* Has been accepted for transmission. */ - if (!asoc->peer.prsctp_capable) - chunk->msg->can_abandon = 0; sctp_chunk_assign_tsn(chunk); - sctp_chunk_assign_ssn(chunk); + asoc->stream.si->assign_number(chunk); } -static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, - struct sctp_chunk *chunk, - u16 chunk_len) +static enum sctp_xmit sctp_packet_will_fit(struct sctp_packet *packet, + struct sctp_chunk *chunk, + u16 chunk_len) { - size_t psize; - size_t pmtu; - int too_big; - sctp_xmit_t retval = SCTP_XMIT_OK; + enum sctp_xmit retval = SCTP_XMIT_OK; + size_t psize, pmtu, maxsize; - psize = packet->size; - pmtu = ((packet->transport->asoc) ? - (packet->transport->asoc->pathmtu) : - (packet->transport->pathmtu)); + /* Don't bundle in this packet if this chunk's auth key doesn't + * match other chunks already enqueued on this packet. Also, + * don't bundle the chunk with auth key if other chunks in this + * packet don't have auth key. + */ + if ((packet->auth && chunk->shkey != packet->auth->shkey) || + (!packet->auth && chunk->shkey && + chunk->chunk_hdr->type != SCTP_CID_AUTH)) + return SCTP_XMIT_PMTU_FULL; - too_big = (psize + chunk_len > pmtu); + psize = packet->size; + if (packet->transport->asoc) + pmtu = packet->transport->asoc->pathmtu; + else + pmtu = packet->transport->pathmtu; /* Decide if we need to fragment or resubmit later. */ - if (too_big) { - /* It's OK to fragmet at IP level if any one of the following + if (psize + chunk_len > pmtu) { + /* It's OK to fragment at IP level if any one of the following * is true: - * 1. The packet is empty (meaning this chunk is greater - * the MTU) - * 2. The chunk we are adding is a control chunk - * 3. The packet doesn't have any data in it yet and data - * requires authentication. + * 1. The packet is empty (meaning this chunk is greater + * the MTU) + * 2. The packet doesn't have any data in it yet and data + * requires authentication. */ - if (sctp_packet_empty(packet) || !sctp_chunk_is_data(chunk) || + if (sctp_packet_empty(packet) || (!packet->has_data && chunk->auth)) { /* We no longer do re-fragmentation. * Just fragment at the IP layer, if we * actually hit this condition */ packet->ipfragok = 1; - } else { - retval = SCTP_XMIT_PMTU_FULL; + goto out; } + + /* Similarly, if this chunk was built before a PMTU + * reduction, we have to fragment it at IP level now. So + * if the packet already contains something, we need to + * flush. + */ + maxsize = pmtu - packet->overhead; + if (packet->auth) + maxsize -= SCTP_PAD4(packet->auth->skb->len); + if (chunk_len > maxsize) + retval = SCTP_XMIT_PMTU_FULL; + + /* It is also okay to fragment if the chunk we are + * adding is a control chunk, but only if current packet + * is not a GSO one otherwise it causes fragmentation of + * a large frame. So in this case we allow the + * fragmentation by forcing it to be in a new packet. + */ + if (!sctp_chunk_is_data(chunk) && packet->has_data) + retval = SCTP_XMIT_PMTU_FULL; + + if (psize + chunk_len > packet->max_size) + /* Hit GSO/PMTU limit, gotta flush */ + retval = SCTP_XMIT_PMTU_FULL; + + if (!packet->transport->burst_limited && + psize + chunk_len > (packet->transport->cwnd >> 1)) + /* Do not allow a single GSO packet to use more + * than half of cwnd. + */ + retval = SCTP_XMIT_PMTU_FULL; + + if (packet->transport->burst_limited && + psize + chunk_len > (packet->transport->burst_limited >> 1)) + /* Do not allow a single GSO packet to use more + * than half of original cwnd. + */ + retval = SCTP_XMIT_PMTU_FULL; + /* Otherwise it will fit in the GSO packet */ } +out: return retval; } |