dccp_packet_hdr_len(dcb->dccpd_type);
int err, set_ack = 1;
u64 ackno = dp->dccps_gsr;
-
- dccp_inc_seqno(&dp->dccps_gss);
+ /*
+ * Increment GSS here already in case the option code needs it.
+ * Update GSS for real only if option processing below succeeds.
+ */
+ dcb->dccpd_seq = ADD48(dp->dccps_gss, 1);
switch (dcb->dccpd_type) {
case DCCP_PKT_DATA:
case DCCP_PKT_REQUEST:
set_ack = 0;
+ /* Use ISS on the first (non-retransmitted) Request. */
+ if (icsk->icsk_retransmits == 0)
+ dcb->dccpd_seq = dp->dccps_iss;
/* fall through */
case DCCP_PKT_SYNC:
break;
}
- dcb->dccpd_seq = dp->dccps_gss;
-
if (dccp_insert_options(sk, skb)) {
kfree_skb(skb);
return -EPROTO;
/* XXX For now we're using only 48 bits sequence numbers */
dh->dccph_x = 1;
- dp->dccps_awh = dp->dccps_gss;
+ dccp_update_gss(sk, dcb->dccpd_seq);
dccp_hdr_set_seq(dh, dp->dccps_gss);
if (set_ack)
dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), ackno);
case DCCP_PKT_REQUEST:
dccp_hdr_request(skb)->dccph_req_service =
dp->dccps_service;
+ /*
+ * Limit Ack window to ISS <= P.ackno <= GSS, so that
+ * only Responses to Requests we sent are considered.
+ */
+ dp->dccps_awl = dp->dccps_iss;
break;
case DCCP_PKT_RESET:
dccp_hdr_reset(skb)->dccph_reset_code =
DCCP_INC_STATS(DCCP_MIB_OUTSEGS);
- memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
err = icsk->icsk_af_ops->queue_xmit(skb, 0);
return net_xmit_eval(err);
}
return -ENOBUFS;
}
+/**
+ * dccp_determine_ccmps - Find out about CCID-specfic packet-size limits
+ * We only consider the HC-sender CCID for setting the CCMPS (RFC 4340, 14.),
+ * since the RX CCID is restricted to feedback packets (Acks), which are small
+ * in comparison with the data traffic. A value of 0 means "no current CCMPS".
+ */
+static u32 dccp_determine_ccmps(const struct dccp_sock *dp)
+{
+ const struct ccid *tx_ccid = dp->dccps_hc_tx_ccid;
+
+ if (tx_ccid == NULL || tx_ccid->ccid_ops == NULL)
+ return 0;
+ return tx_ccid->ccid_ops->ccid_ccmps;
+}
+
unsigned int dccp_sync_mss(struct sock *sk, u32 pmtu)
{
struct inet_connection_sock *icsk = inet_csk(sk);
struct dccp_sock *dp = dccp_sk(sk);
- int mss_now = (pmtu - icsk->icsk_af_ops->net_header_len -
- sizeof(struct dccp_hdr) - sizeof(struct dccp_hdr_ext));
+ u32 ccmps = dccp_determine_ccmps(dp);
+ u32 cur_mps = ccmps ? min(pmtu, ccmps) : pmtu;
- /* Now subtract optional transport overhead */
- mss_now -= icsk->icsk_ext_hdr_len;
+ /* Account for header lengths and IPv4/v6 option overhead */
+ cur_mps -= (icsk->icsk_af_ops->net_header_len + icsk->icsk_ext_hdr_len +
+ sizeof(struct dccp_hdr) + sizeof(struct dccp_hdr_ext));
/*
- * FIXME: this should come from the CCID infrastructure, where, say,
- * TFRC will say it wants TIMESTAMPS, ELAPSED time, etc, for now lets
- * put a rough estimate for NDP + TIMESTAMP + TIMESTAMP_ECHO + ELAPSED
- * TIME + TFRC_OPT_LOSS_EVENT_RATE + TFRC_OPT_RECEIVE_RATE + padding to
- * make it a multiple of 4
+ * Leave enough headroom for common DCCP header options.
+ * This only considers options which may appear on DCCP-Data packets, as
+ * per table 3 in RFC 4340, 5.8. When running out of space for other
+ * options (eg. Ack Vector which can take up to 255 bytes), it is better
+ * to schedule a separate Ack. Thus we leave headroom for the following:
+ * - 1 byte for Slow Receiver (11.6)
+ * - 6 bytes for Timestamp (13.1)
+ * - 10 bytes for Timestamp Echo (13.3)
+ * - 8 bytes for NDP count (7.7, when activated)
+ * - 6 bytes for Data Checksum (9.3)
+ * - %DCCPAV_MIN_OPTLEN bytes for Ack Vector size (11.4, when enabled)
*/
-
- mss_now -= ((5 + 6 + 10 + 6 + 6 + 6 + 3) / 4) * 4;
+ cur_mps -= roundup(1 + 6 + 10 + dp->dccps_send_ndp_count * 8 + 6 +
+ (dp->dccps_hc_rx_ackvec ? DCCPAV_MIN_OPTLEN : 0), 4);
/* And store cached results */
icsk->icsk_pmtu_cookie = pmtu;
- dp->dccps_mss_cache = mss_now;
+ dp->dccps_mss_cache = cur_mps;
- return mss_now;
+ return cur_mps;
}
EXPORT_SYMBOL_GPL(dccp_sync_mss);
const int len = skb->len;
if (sk->sk_state == DCCP_PARTOPEN) {
- /* See 8.1.5. Handshake Completion */
+ const u32 cur_mps = dp->dccps_mss_cache - DCCP_FEATNEG_OVERHEAD;
+ /*
+ * See 8.1.5 - Handshake Completion.
+ *
+ * For robustness we resend Confirm options until the client has
+ * entered OPEN. During the initial feature negotiation, the MPS
+ * is smaller than usual, reduced by the Change/Confirm options.
+ */
+ if (!list_empty(&dp->dccps_featneg) && len > cur_mps) {
+ DCCP_WARN("Payload too large (%d) for featneg.\n", len);
+ dccp_send_ack(sk);
+ dccp_feat_list_purge(&dp->dccps_featneg);
+ }
+
inet_csk_schedule_ack(sk);
inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK,
inet_csk(sk)->icsk_rto,
}
}
-int dccp_retransmit_skb(struct sock *sk, struct sk_buff *skb)
+/**
+ * dccp_retransmit_skb - Retransmit Request, Close, or CloseReq packets
+ * There are only four retransmittable packet types in DCCP:
+ * - Request in client-REQUEST state (sec. 8.1.1),
+ * - CloseReq in server-CLOSEREQ state (sec. 8.3),
+ * - Close in node-CLOSING state (sec. 8.3),
+ * - Acks in client-PARTOPEN state (sec. 8.1.5, handled by dccp_delack_timer()).
+ * This function expects sk->sk_send_head to contain the original skb.
+ */
+int dccp_retransmit_skb(struct sock *sk)
{
+ WARN_ON(sk->sk_send_head == NULL);
+
if (inet_csk(sk)->icsk_af_ops->rebuild_header(sk) != 0)
return -EHOSTUNREACH; /* Routing failure or similar. */
- return dccp_transmit_skb(sk, (skb_cloned(skb) ?
- pskb_copy(skb, GFP_ATOMIC):
- skb_clone(skb, GFP_ATOMIC)));
+ /* this count is used to distinguish original and retransmitted skb */
+ inet_csk(sk)->icsk_retransmits++;
+
+ return dccp_transmit_skb(sk, skb_clone(sk->sk_send_head, GFP_ATOMIC));
}
struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst,
/* Reserve space for headers. */
skb_reserve(skb, sk->sk_prot->max_header);
- skb->dst = dst_clone(dst);
+ skb_dst_set(skb, dst_clone(dst));
dreq = dccp_rsk(req);
if (inet_rsk(req)->acked) /* increase ISS upon retransmission */
DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_RESPONSE;
DCCP_SKB_CB(skb)->dccpd_seq = dreq->dreq_iss;
- if (dccp_insert_options(sk, skb)) {
- kfree_skb(skb);
- return NULL;
- }
+ /* Resolve feature dependencies resulting from choice of CCID */
+ if (dccp_feat_server_ccid_dependencies(dreq))
+ goto response_failed;
+
+ if (dccp_insert_options_rsk(dreq, skb))
+ goto response_failed;
/* Build and checksum header */
dh = dccp_zeroed_hdr(skb, dccp_header_size);
- dh->dccph_sport = inet_sk(sk)->sport;
+ dh->dccph_sport = inet_rsk(req)->loc_port;
dh->dccph_dport = inet_rsk(req)->rmt_port;
dh->dccph_doff = (dccp_header_size +
DCCP_SKB_CB(skb)->dccpd_opt_len) / 4;
inet_rsk(req)->acked = 1;
DCCP_INC_STATS(DCCP_MIB_OUTSEGS);
return skb;
+response_failed:
+ kfree_skb(skb);
+ return NULL;
}
EXPORT_SYMBOL_GPL(dccp_make_response);
/* answer offending packet in @rcv_skb with Reset from control socket @ctl */
-struct sk_buff *dccp_ctl_make_reset(struct socket *ctl, struct sk_buff *rcv_skb)
+struct sk_buff *dccp_ctl_make_reset(struct sock *sk, struct sk_buff *rcv_skb)
{
struct dccp_hdr *rxdh = dccp_hdr(rcv_skb), *dh;
struct dccp_skb_cb *dcb = DCCP_SKB_CB(rcv_skb);
struct dccp_hdr_reset *dhr;
struct sk_buff *skb;
- skb = alloc_skb(ctl->sk->sk_prot->max_header, GFP_ATOMIC);
+ skb = alloc_skb(sk->sk_prot->max_header, GFP_ATOMIC);
if (skb == NULL)
return NULL;
- skb_reserve(skb, ctl->sk->sk_prot->max_header);
+ skb_reserve(skb, sk->sk_prot->max_header);
/* Swap the send and the receive. */
dh = dccp_zeroed_hdr(skb, dccp_hdr_reset_len);
dccp_sync_mss(sk, dst_mtu(dst));
- /*
- * SWL and AWL are initially adjusted so that they are not less than
- * the initial Sequence Numbers received and sent, respectively:
- * SWL := max(GSR + 1 - floor(W/4), ISR),
- * AWL := max(GSS - W' + 1, ISS).
- * These adjustments MUST be applied only at the beginning of the
- * connection.
- */
- dccp_update_gss(sk, dp->dccps_iss);
- dccp_set_seqno(&dp->dccps_awl, max48(dp->dccps_awl, dp->dccps_iss));
-
- /* S.GAR - greatest valid acknowledgement number received on a non-Sync;
- * initialized to S.ISS (sec. 8.5) */
+ /* Initialise GAR as per 8.5; AWL/AWH are set in dccp_transmit_skb() */
dp->dccps_gar = dp->dccps_iss;
icsk->icsk_retransmits = 0;
struct sk_buff *skb;
struct inet_connection_sock *icsk = inet_csk(sk);
+ /* do not connect if feature negotiation setup fails */
+ if (dccp_feat_finalise_settings(dccp_sk(sk)))
+ return -EPROTO;
+
dccp_connect_init(sk);
skb = alloc_skb(sk->sk_prot->max_header, sk->sk_allocation);
EXPORT_SYMBOL_GPL(dccp_send_ack);
+#if 0
/* FIXME: Is this still necessary (11.3) - currently nowhere used by DCCP. */
void dccp_send_delayed_ack(struct sock *sk)
{
icsk->icsk_ack.timeout = timeout;
sk_reset_timer(sk, &icsk->icsk_delack_timer, timeout);
}
+#endif
void dccp_send_sync(struct sock *sk, const u64 ackno,
const enum dccp_pkt_type pkt_type)
/* Reserve space for headers and prepare control bits. */
skb_reserve(skb, sk->sk_prot->max_header);
- DCCP_SKB_CB(skb)->dccpd_type = dp->dccps_role == DCCP_ROLE_CLIENT ?
- DCCP_PKT_CLOSE : DCCP_PKT_CLOSEREQ;
+ if (dp->dccps_role == DCCP_ROLE_SERVER && !dp->dccps_server_timewait)
+ DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_CLOSEREQ;
+ else
+ DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_CLOSE;
if (active) {
dccp_write_xmit(sk, 1);
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff