Merge branch 'dccp' of git://eden-feed.erg.abdn.ac.uk/dccp_exp

Conflicts:

	net/dccp/input.c
	net/dccp/options.c

Documentation/networking/dccp.txt

@@ -45,6 +45,25 @@ http://linux-net.osdl.org/index.php/DCCP_Testing#Experimental_DCCP_source_tree
 
 Socket options
 ==============
+DCCP_SOCKOPT_QPOLICY_ID sets the dequeuing policy for outgoing packets. It takes
+a policy ID as argument and can only be set before the connection (i.e. changes
+during an established connection are not supported). Currently, two policies are
+defined: the "simple" policy (DCCPQ_POLICY_SIMPLE), which does nothing special,
+and a priority-based variant (DCCPQ_POLICY_PRIO). The latter allows to pass an
+u32 priority value as ancillary data to sendmsg(), where higher numbers indicate
+a higher packet priority (similar to SO_PRIORITY). This ancillary data needs to
+be formatted using a cmsg(3) message header filled in as follows:
+	cmsg->cmsg_level = SOL_DCCP;
+	cmsg->cmsg_type	 = DCCP_SCM_PRIORITY;
+	cmsg->cmsg_len	 = CMSG_LEN(sizeof(uint32_t));	/* or CMSG_LEN(4) */
+
+DCCP_SOCKOPT_QPOLICY_TXQLEN sets the maximum length of the output queue. A zero
+value is always interpreted as unbounded queue length. If different from zero,
+the interpretation of this parameter depends on the current dequeuing policy
+(see above): the "simple" policy will enforce a fixed queue size by returning
+EAGAIN, whereas the "prio" policy enforces a fixed queue length by dropping the
+lowest-priority packet first. The default value for this parameter is
+initialised from /proc/sys/net/dccp/default/tx_qlen.
 
 DCCP_SOCKOPT_SERVICE sets the service. The specification mandates use of
 service codes (RFC 4340, sec. 8.1.2); if this socket option is not set,
@@ -57,6 +76,24 @@ can be set before calling bind().
 DCCP_SOCKOPT_GET_CUR_MPS is read-only and retrieves the current maximum packet
 size (application payload size) in bytes, see RFC 4340, section 14.
 
+DCCP_SOCKOPT_AVAILABLE_CCIDS is also read-only and returns the list of CCIDs
+supported by the endpoint (see include/linux/dccp.h for symbolic constants).
+The caller needs to provide a sufficiently large (> 2) array of type uint8_t.
+
+DCCP_SOCKOPT_CCID is write-only and sets both the TX and RX CCIDs at the same
+time, combining the operation of the next two socket options. This option is
+preferrable over the latter two, since often applications will use the same
+type of CCID for both directions; and mixed use of CCIDs is not currently well
+understood. This socket option takes as argument at least one uint8_t value, or
+an array of uint8_t values, which must match available CCIDS (see above). CCIDs
+must be registered on the socket before calling connect() or listen().
+
+DCCP_SOCKOPT_TX_CCID is read/write. It returns the current CCID (if set) or sets
+the preference list for the TX CCID, using the same format as DCCP_SOCKOPT_CCID.
+Please note that the getsockopt argument type here is `int', not uint8_t.
+
+DCCP_SOCKOPT_RX_CCID is analogous to DCCP_SOCKOPT_TX_CCID, but for the RX CCID.
+
 DCCP_SOCKOPT_SERVER_TIMEWAIT enables the server (listening socket) to hold
 timewait state when closing the connection (RFC 4340, 8.3). The usual case is
 that the closing server sends a CloseReq, whereupon the client holds timewait
@@ -115,23 +152,16 @@ retries2
 	importance for retransmitted acknowledgments and feature negotiation,
 	data packets are never retransmitted. Analogue of tcp_retries2.
 
-send_ndp = 1
-	Whether or not to send NDP count options (sec. 7.7.2).
-
-send_ackvec = 1
-	Whether or not to send Ack Vector options (sec. 11.5).
-
-ack_ratio = 2
-	The default Ack Ratio (sec. 11.3) to use.
-
 tx_ccid = 2
-	Default CCID for the sender-receiver half-connection.
+	Default CCID for the sender-receiver half-connection. Depending on the
+	choice of CCID, the Send Ack Vector feature is enabled automatically.
 
 rx_ccid = 2
-	Default CCID for the receiver-sender half-connection.
+	Default CCID for the receiver-sender half-connection; see tx_ccid.
 
 seq_window = 100
-	The initial sequence window (sec. 7.5.2).
+	The initial sequence window (sec. 7.5.2) of the sender. This influences
+	the local ackno validity and the remote seqno validity windows (7.5.1).
 
 tx_qlen = 5
 	The size of the transmit buffer in packets. A value of 0 corresponds

include/linux/dccp.h

@@ -165,9 +165,13 @@ enum {
 	DCCPO_TIMESTAMP_ECHO = 42,
 	DCCPO_ELAPSED_TIME = 43,
 	DCCPO_MAX = 45,
-	DCCPO_MIN_CCID_SPECIFIC = 128,
-	DCCPO_MAX_CCID_SPECIFIC = 255,
+	DCCPO_MIN_RX_CCID_SPECIFIC = 128,	/* from sender to receiver */
+	DCCPO_MAX_RX_CCID_SPECIFIC = 191,
+	DCCPO_MIN_TX_CCID_SPECIFIC = 192,	/* from receiver to sender */
+	DCCPO_MAX_TX_CCID_SPECIFIC = 255,
 };
+/* maximum size of a single TLV-encoded DCCP option (sans type/len bytes) */
+#define DCCP_SINGLE_OPT_MAXLEN	253
 
 /* DCCP CCIDS */
 enum {
@@ -176,27 +180,36 @@ enum {
 };
 
 /* DCCP features (RFC 4340 section 6.4) */
-enum {
+enum dccp_feature_numbers {
 	DCCPF_RESERVED = 0,
 	DCCPF_CCID = 1,
-	DCCPF_SHORT_SEQNOS = 2,		/* XXX: not yet implemented */
+	DCCPF_SHORT_SEQNOS = 2,
 	DCCPF_SEQUENCE_WINDOW = 3,
-	DCCPF_ECN_INCAPABLE = 4,	/* XXX: not yet implemented */
+	DCCPF_ECN_INCAPABLE = 4,
 	DCCPF_ACK_RATIO = 5,
 	DCCPF_SEND_ACK_VECTOR = 6,
 	DCCPF_SEND_NDP_COUNT = 7,
 	DCCPF_MIN_CSUM_COVER = 8,
-	DCCPF_DATA_CHECKSUM = 9,	/* XXX: not yet implemented */
+	DCCPF_DATA_CHECKSUM = 9,
 	/* 10-127 reserved */
 	DCCPF_MIN_CCID_SPECIFIC = 128,
+	DCCPF_SEND_LEV_RATE = 192,	/* RFC 4342, sec. 8.4 */
 	DCCPF_MAX_CCID_SPECIFIC = 255,
 };
 
-/* this structure is argument to DCCP_SOCKOPT_CHANGE_X */
-struct dccp_so_feat {
-	__u8 dccpsf_feat;
-	__u8 __user *dccpsf_val;
-	__u8 dccpsf_len;
+/* DCCP socket control message types for cmsg */
+enum dccp_cmsg_type {
+	DCCP_SCM_PRIORITY = 1,
+	DCCP_SCM_QPOLICY_MAX = 0xFFFF,
+	/* ^-- Up to here reserved exclusively for qpolicy parameters */
+	DCCP_SCM_MAX
+};
+
+/* DCCP priorities for outgoing/queued packets */
+enum dccp_packet_dequeueing_policy {
+	DCCPQ_POLICY_SIMPLE,
+	DCCPQ_POLICY_PRIO,
+	DCCPQ_POLICY_MAX
 };
 
 /* DCCP socket options */
@@ -208,6 +221,12 @@ struct dccp_so_feat {
 #define DCCP_SOCKOPT_SERVER_TIMEWAIT	6
 #define DCCP_SOCKOPT_SEND_CSCOV		10
 #define DCCP_SOCKOPT_RECV_CSCOV		11
+#define DCCP_SOCKOPT_AVAILABLE_CCIDS	12
+#define DCCP_SOCKOPT_CCID		13
+#define DCCP_SOCKOPT_TX_CCID		14
+#define DCCP_SOCKOPT_RX_CCID		15
+#define DCCP_SOCKOPT_QPOLICY_ID		16
+#define DCCP_SOCKOPT_QPOLICY_TXQLEN	17
 #define DCCP_SOCKOPT_CCID_RX_INFO	128
 #define DCCP_SOCKOPT_CCID_TX_INFO	192
 
@@ -355,62 +374,13 @@ static inline unsigned int dccp_hdr_len(const struct sk_buff *skb)
 	return __dccp_hdr_len(dccp_hdr(skb));
 }
 
-
-/* initial values for each feature */
-#define DCCPF_INITIAL_SEQUENCE_WINDOW		100
-#define DCCPF_INITIAL_ACK_RATIO			2
-#define DCCPF_INITIAL_CCID			DCCPC_CCID2
-#define DCCPF_INITIAL_SEND_ACK_VECTOR		1
-/* FIXME: for now we're default to 1 but it should really be 0 */
-#define DCCPF_INITIAL_SEND_NDP_COUNT		1
-
-/**
-  * struct dccp_minisock - Minimal DCCP connection representation
-  *
-  * Will be used to pass the state from dccp_request_sock to dccp_sock.
-  *
-  * @dccpms_sequence_window - Sequence Window Feature (section 7.5.2)
-  * @dccpms_ccid - Congestion Control Id (CCID) (section 10)
-  * @dccpms_send_ack_vector - Send Ack Vector Feature (section 11.5)
-  * @dccpms_send_ndp_count - Send NDP Count Feature (7.7.2)
-  * @dccpms_ack_ratio - Ack Ratio Feature (section 11.3)
-  * @dccpms_pending - List of features being negotiated
-  * @dccpms_conf -
-  */
-struct dccp_minisock {
-	__u64			dccpms_sequence_window;
-	__u8			dccpms_rx_ccid;
-	__u8			dccpms_tx_ccid;
-	__u8			dccpms_send_ack_vector;
-	__u8			dccpms_send_ndp_count;
-	__u8			dccpms_ack_ratio;
-	struct list_head	dccpms_pending;
-	struct list_head	dccpms_conf;
-};
-
-struct dccp_opt_conf {
-	__u8			*dccpoc_val;
-	__u8			dccpoc_len;
-};
-
-struct dccp_opt_pend {
-	struct list_head	dccpop_node;
-	__u8			dccpop_type;
-	__u8			dccpop_feat;
-	__u8		        *dccpop_val;
-	__u8			dccpop_len;
-	int			dccpop_conf;
-	struct dccp_opt_conf    *dccpop_sc;
-};
-
-extern void dccp_minisock_init(struct dccp_minisock *dmsk);
-
 /**
  * struct dccp_request_sock  -  represent DCCP-specific connection request
  * @dreq_inet_rsk: structure inherited from
  * @dreq_iss: initial sequence number sent on the Response (RFC 4340, 7.1)
  * @dreq_isr: initial sequence number received on the Request
  * @dreq_service: service code present on the Request (there is just one)
+ * @dreq_featneg: feature negotiation options for this connection
  * The following two fields are analogous to the ones in dccp_sock:
  * @dreq_timestamp_echo: last received timestamp to echo (13.1)
  * @dreq_timestamp_echo: the time of receiving the last @dreq_timestamp_echo
@@ -420,6 +390,7 @@ struct dccp_request_sock {
 	__u64			 dreq_iss;
 	__u64			 dreq_isr;
 	__be32			 dreq_service;
+	struct list_head	 dreq_featneg;
 	__u32			 dreq_timestamp_echo;
 	__u32			 dreq_timestamp_time;
 };
@@ -491,21 +462,28 @@ struct dccp_ackvec;
  * @dccps_timestamp_time - time of receiving latest @dccps_timestamp_echo
  * @dccps_l_ack_ratio - feature-local Ack Ratio
  * @dccps_r_ack_ratio - feature-remote Ack Ratio
+ * @dccps_l_seq_win - local Sequence Window (influences ack number validity)
+ * @dccps_r_seq_win - remote Sequence Window (influences seq number validity)
  * @dccps_pcslen - sender   partial checksum coverage (via sockopt)
  * @dccps_pcrlen - receiver partial checksum coverage (via sockopt)
+ * @dccps_send_ndp_count - local Send NDP Count feature (7.7.2)
  * @dccps_ndp_count - number of Non Data Packets since last data packet
  * @dccps_mss_cache - current value of MSS (path MTU minus header sizes)
  * @dccps_rate_last - timestamp for rate-limiting DCCP-Sync (RFC 4340, 7.5.4)
- * @dccps_minisock - associated minisock (accessed via dccp_msk)
+ * @dccps_featneg - tracks feature-negotiation state (mostly during handshake)
  * @dccps_hc_rx_ackvec - rx half connection ack vector
  * @dccps_hc_rx_ccid - CCID used for the receiver (or receiving half-connection)
  * @dccps_hc_tx_ccid - CCID used for the sender (or sending half-connection)
  * @dccps_options_received - parsed set of retrieved options
+ * @dccps_qpolicy - TX dequeueing policy, one of %dccp_packet_dequeueing_policy
+ * @dccps_tx_qlen - maximum length of the TX queue
  * @dccps_role - role of this sock, one of %dccp_role
  * @dccps_hc_rx_insert_options - receiver wants to add options when acking
  * @dccps_hc_tx_insert_options - sender wants to add options when sending
  * @dccps_server_timewait - server holds timewait state on close (RFC 4340, 8.3)
- * @dccps_xmit_timer - timer for when CCID is not ready to send
+ * @dccps_sync_scheduled - flag which signals "send out-of-band message soon"
+ * @dccps_xmitlet - tasklet scheduled by the TX CCID to dequeue data packets
+ * @dccps_xmit_timer - used by the TX CCID to delay sending (rate-based pacing)
  * @dccps_syn_rtt - RTT sample from Request/Response exchange (in usecs)
  */
 struct dccp_sock {
@@ -529,19 +507,26 @@ struct dccp_sock {
 	__u32				dccps_timestamp_time;
 	__u16				dccps_l_ack_ratio;
 	__u16				dccps_r_ack_ratio;
-	__u16				dccps_pcslen;
-	__u16				dccps_pcrlen;
+	__u64				dccps_l_seq_win:48;
+	__u64				dccps_r_seq_win:48;
+	__u8				dccps_pcslen:4;
+	__u8				dccps_pcrlen:4;
+	__u8				dccps_send_ndp_count:1;
 	__u64				dccps_ndp_count:48;
 	unsigned long			dccps_rate_last;
-	struct dccp_minisock		dccps_minisock;
+	struct list_head		dccps_featneg;
 	struct dccp_ackvec		*dccps_hc_rx_ackvec;
 	struct ccid			*dccps_hc_rx_ccid;
 	struct ccid			*dccps_hc_tx_ccid;
 	struct dccp_options_received	dccps_options_received;
+	__u8				dccps_qpolicy;
+	__u32				dccps_tx_qlen;
 	enum dccp_role			dccps_role:2;
 	__u8				dccps_hc_rx_insert_options:1;
 	__u8				dccps_hc_tx_insert_options:1;
 	__u8				dccps_server_timewait:1;
+	__u8				dccps_sync_scheduled:1;
+	struct tasklet_struct		dccps_xmitlet;
 	struct timer_list		dccps_xmit_timer;
 };
 
@@ -550,11 +535,6 @@ static inline struct dccp_sock *dccp_sk(const struct sock *sk)
 	return (struct dccp_sock *)sk;
 }
 
-static inline struct dccp_minisock *dccp_msk(const struct sock *sk)
-{
-	return (struct dccp_minisock *)&dccp_sk(sk)->dccps_minisock;
-}
-
 static inline const char *dccp_role(const struct sock *sk)
 {
 	switch (dccp_sk(sk)->dccps_role) {

include/net/tcp.h

@@ -782,6 +782,21 @@ static inline __u32 tcp_current_ssthresh(const struct sock *sk)
 /* Use define here intentionally to get WARN_ON location shown at the caller */
 #define tcp_verify_left_out(tp)	WARN_ON(tcp_left_out(tp) > tp->packets_out)
 
+/*
+ * Convert RFC3390 larger initial windows into an equivalent number of packets.
+ *
+ * John Heffner states:
+ *
+ *	The RFC specifies a window of no more than 4380 bytes
+ *	unless 2*MSS > 4380.  Reading the pseudocode in the RFC
+ *	is a bit misleading because they use a clamp at 4380 bytes
+ *	rather than a multiplier in the relevant range.
+ */
+static inline u32 rfc3390_bytes_to_packets(const u32 bytes)
+{
+	return bytes <= 1095 ? 4 : (bytes > 1460 ? 2 : 3);
+}
+
 extern void tcp_enter_cwr(struct sock *sk, const int set_ssthresh);
 extern __u32 tcp_init_cwnd(struct tcp_sock *tp, struct dst_entry *dst);
 

net/dccp/Kconfig

@@ -25,9 +25,6 @@ config INET_DCCP_DIAG
 	def_tristate y if (IP_DCCP = y && INET_DIAG = y)
 	def_tristate m
 
-config IP_DCCP_ACKVEC
-	bool
-
 source "net/dccp/ccids/Kconfig"
 
 menu "DCCP Kernel Hacking"

net/dccp/Makefile

 obj-$(CONFIG_IP_DCCP) += dccp.o dccp_ipv4.o
 
-dccp-y := ccid.o feat.o input.o minisocks.o options.o output.o proto.o timer.o
+dccp-y := ccid.o feat.o input.o minisocks.o options.o \
+	  qpolicy.o output.o proto.o timer.o ackvec.o
 
 dccp_ipv4-y := ipv4.o
 
@@ -8,8 +9,6 @@ dccp_ipv4-y := ipv4.o
 obj-$(subst y,$(CONFIG_IP_DCCP),$(CONFIG_IPV6)) += dccp_ipv6.o
 dccp_ipv6-y := ipv6.o
 
-dccp-$(CONFIG_IP_DCCP_ACKVEC) += ackvec.o
-
 obj-$(CONFIG_INET_DCCP_DIAG) += dccp_diag.o
 obj-$(CONFIG_NET_DCCPPROBE) += dccp_probe.o
 

net/dccp/ackvec.h

@@ -3,156 +3,134 @@
 /*
  *  net/dccp/ackvec.h
  *
- *  An implementation of the DCCP protocol
+ *  An implementation of Ack Vectors for the DCCP protocol
+ *  Copyright (c) 2007 University of Aberdeen, Scotland, UK
  *  Copyright (c) 2005 Arnaldo Carvalho de Melo <acme@mandriva.com>
- *
  *	This program is free software; you can redistribute it and/or modify it
  *	under the terms of the GNU General Public License version 2 as
  *	published by the Free Software Foundation.
  */
 
+#include <linux/dccp.h>
 #include <linux/compiler.h>
-#include <linux/ktime.h>
 #include <linux/list.h>
 #include <linux/types.h>
 
-/* Read about the ECN nonce to see why it is 253 */
-#define DCCP_MAX_ACKVEC_OPT_LEN 253
-/* We can spread an ack vector across multiple options */
-#define DCCP_MAX_ACKVEC_LEN (DCCP_MAX_ACKVEC_OPT_LEN * 2)
+/*
+ * Ack Vector buffer space is static, in multiples of %DCCP_SINGLE_OPT_MAXLEN,
+ * the maximum size of a single Ack Vector. Setting %DCCPAV_NUM_ACKVECS to 1
+ * will be sufficient for most cases of low Ack Ratios, using a value of 2 gives
+ * more headroom if Ack Ratio is higher or when the sender acknowledges slowly.
+ * The maximum value is bounded by the u16 types for indices and functions.
+ */
+#define DCCPAV_NUM_ACKVECS	2
+#define DCCPAV_MAX_ACKVEC_LEN	(DCCP_SINGLE_OPT_MAXLEN * DCCPAV_NUM_ACKVECS)
 
-#define DCCP_ACKVEC_STATE_RECEIVED	0
-#define DCCP_ACKVEC_STATE_ECN_MARKED	(1 << 6)
-#define DCCP_ACKVEC_STATE_NOT_RECEIVED	(3 << 6)
+/* Estimated minimum average Ack Vector length - used for updating MPS */
+#define DCCPAV_MIN_OPTLEN	16
 
-#define DCCP_ACKVEC_STATE_MASK		0xC0 /* 11000000 */
-#define DCCP_ACKVEC_LEN_MASK		0x3F /* 00111111 */
+/* Threshold for coping with large bursts of losses */
+#define DCCPAV_BURST_THRESH	(DCCPAV_MAX_ACKVEC_LEN / 8)
 
-/** struct dccp_ackvec - ack vector
- *
- * This data structure is the one defined in RFC 4340, Appendix A.
- *
- * @av_buf_head - circular buffer head
- * @av_buf_tail - circular buffer tail
- * @av_buf_ackno - ack # of the most recent packet acknowledgeable in the
- *		       buffer (i.e. %av_buf_head)
- * @av_buf_nonce - the one-bit sum of the ECN Nonces on all packets acked
- * 		       by the buffer with State 0
- *
- * Additionally, the HC-Receiver must keep some information about the
- * Ack Vectors it has recently sent. For each packet sent carrying an
- * Ack Vector, it remembers four variables:
+enum dccp_ackvec_states {
+	DCCPAV_RECEIVED =	0x00,
+	DCCPAV_ECN_MARKED =	0x40,
+	DCCPAV_RESERVED =	0x80,
+	DCCPAV_NOT_RECEIVED =	0xC0
+};
+#define DCCPAV_MAX_RUNLEN	0x3F
+
+static inline u8 dccp_ackvec_runlen(const u8 *cell)
+{
+	return *cell & DCCPAV_MAX_RUNLEN;
+}
+
+static inline u8 dccp_ackvec_state(const u8 *cell)
+{
+	return *cell & ~DCCPAV_MAX_RUNLEN;
+}
+
+/** struct dccp_ackvec - Ack Vector main data structure
  *
- * @av_records - list of dccp_ackvec_record
- * @av_ack_nonce - the one-bit sum of the ECN Nonces for all State 0.
+ * This implements a fixed-size circular buffer within an array and is largely
+ * based on Appendix A of RFC 4340.
  *
- * @av_time - the time in usecs
- * @av_buf - circular buffer of acknowledgeable packets
+ * @av_buf:	   circular buffer storage area
+ * @av_buf_head:   head index; begin of live portion in @av_buf
+ * @av_buf_tail:   tail index; first index _after_ the live portion in @av_buf
+ * @av_buf_ackno:  highest seqno of acknowledgeable packet recorded in @av_buf
+ * @av_tail_ackno: lowest  seqno of acknowledgeable packet recorded in @av_buf
+ * @av_buf_nonce:  ECN nonce sums, each covering subsequent segments of up to
+ *		   %DCCP_SINGLE_OPT_MAXLEN cells in the live portion of @av_buf
+ * @av_overflow:   if 1 then buf_head == buf_tail indicates buffer wraparound
+ * @av_records:	   list of %dccp_ackvec_record (Ack Vectors sent previously)
  */
 struct dccp_ackvec {
-	u64			av_buf_ackno;
-	struct list_head	av_records;
-	ktime_t			av_time;
+	u8			av_buf[DCCPAV_MAX_ACKVEC_LEN];
 	u16			av_buf_head;
-	u16			av_vec_len;
-	u8			av_buf_nonce;
-	u8			av_ack_nonce;
-	u8			av_buf[DCCP_MAX_ACKVEC_LEN];
+	u16			av_buf_tail;
+	u64			av_buf_ackno:48;
+	u64			av_tail_ackno:48;
+	bool			av_buf_nonce[DCCPAV_NUM_ACKVECS];
+	u8			av_overflow:1;
+	struct list_head	av_records;
 };
 
-/** struct dccp_ackvec_record - ack vector record
+/** struct dccp_ackvec_record - Records information about sent Ack Vectors
  *
- * ACK vector record as defined in Appendix A of spec.
+ * These list entries define the additional information which the HC-Receiver
+ * keeps about recently-sent Ack Vectors; again refer to RFC 4340, Appendix A.
  *
- * The list is sorted by avr_ack_seqno
+ * @avr_node:	    the list node in @av_records
+ * @avr_ack_seqno:  sequence number of the packet the Ack Vector was sent on
+ * @avr_ack_ackno:  the Ack number that this record/Ack Vector refers to
+ * @avr_ack_ptr:    pointer into @av_buf where this record starts
+ * @avr_ack_runlen: run length of @avr_ack_ptr at the time of sending
+ * @avr_ack_nonce:  the sum of @av_buf_nonce's at the time this record was sent
  *
- * @avr_node - node in av_records
- * @avr_ack_seqno - sequence number of the packet this record was sent on
- * @avr_ack_ackno - sequence number being acknowledged
- * @avr_ack_ptr - pointer into av_buf where this record starts
- * @avr_ack_nonce - av_ack_nonce at the time this record was sent
- * @avr_sent_len - lenght of the record in av_buf
+ * The list as a whole is sorted in descending order by @avr_ack_seqno.
  */
 struct dccp_ackvec_record {
 	struct list_head avr_node;
-	u64		 avr_ack_seqno;
-	u64		 avr_ack_ackno;
+	u64		 avr_ack_seqno:48;
+	u64		 avr_ack_ackno:48;
 	u16		 avr_ack_ptr;
-	u16		 avr_sent_len;
-	u8		 avr_ack_nonce;
+	u8		 avr_ack_runlen;
+	u8		 avr_ack_nonce:1;
 };
 
-struct sock;
-struct sk_buff;
-
-#ifdef CONFIG_IP_DCCP_ACKVEC
-extern int dccp_ackvec_init(void);
+extern int  dccp_ackvec_init(void);
 extern void dccp_ackvec_exit(void);
 
 extern struct dccp_ackvec *dccp_ackvec_alloc(const gfp_t priority);
 extern void dccp_ackvec_free(struct dccp_ackvec *av);
 
-extern int dccp_ackvec_add(struct dccp_ackvec *av, const struct sock *sk,
-			   const u64 ackno, const u8 state);
-
-extern void dccp_ackvec_check_rcv_ackno(struct dccp_ackvec *av,
-					struct sock *sk, const u64 ackno);
-extern int dccp_ackvec_parse(struct sock *sk, const struct sk_buff *skb,
-			     u64 *ackno, const u8 opt,
-			     const u8 *value, const u8 len);
+extern void dccp_ackvec_input(struct dccp_ackvec *av, struct sk_buff *skb);
+extern int  dccp_ackvec_update_records(struct dccp_ackvec *av, u64 seq, u8 sum);
+extern void dccp_ackvec_clear_state(struct dccp_ackvec *av, const u64 ackno);
+extern u16  dccp_ackvec_buflen(const struct dccp_ackvec *av);
 
-extern int dccp_insert_option_ackvec(struct sock *sk, struct sk_buff *skb);
-
-static inline int dccp_ackvec_pending(const struct dccp_ackvec *av)
-{
-	return av->av_vec_len;
-}
-#else /* CONFIG_IP_DCCP_ACKVEC */
-static inline int dccp_ackvec_init(void)
+static inline bool dccp_ackvec_is_empty(const struct dccp_ackvec *av)
 {
-	return 0;
+	return av->av_overflow == 0 && av->av_buf_head == av->av_buf_tail;
 }
 
-static inline void dccp_ackvec_exit(void)
-{
-}
-
-static inline struct dccp_ackvec *dccp_ackvec_alloc(const gfp_t priority)
-{
-	return NULL;
-}
-
-static inline void dccp_ackvec_free(struct dccp_ackvec *av)
-{
-}
-
-static inline int dccp_ackvec_add(struct dccp_ackvec *av, const struct sock *sk,
-				  const u64 ackno, const u8 state)
-{
-	return -1;
-}
-
-static inline void dccp_ackvec_check_rcv_ackno(struct dccp_ackvec *av,
-					       struct sock *sk, const u64 ackno)
-{
-}
-
-static inline int dccp_ackvec_parse(struct sock *sk, const struct sk_buff *skb,
-				    const u64 *ackno, const u8 opt,
-				    const u8 *value, const u8 len)
-{
-	return -1;
-}
-
-static inline int dccp_insert_option_ackvec(const struct sock *sk,
-					    const struct sk_buff *skb)
-{
-	return -1;
-}
+/**
+ * struct dccp_ackvec_parsed  -  Record offsets of Ack Vectors in skb
+ * @vec:	start of vector (offset into skb)
+ * @len:	length of @vec
+ * @nonce:	whether @vec had an ECN nonce of 0 or 1
+ * @node:	FIFO - arranged in descending order of ack_ackno
+ * This structure is used by CCIDs to access Ack Vectors in a received skb.
+ */
+struct dccp_ackvec_parsed {
+	u8		 *vec,
+			 len,
+			 nonce:1;
+	struct list_head node;
+};
 
-static inline int dccp_ackvec_pending(const struct dccp_ackvec *av)
-{
-	return 0;
-}
-#endif /* CONFIG_IP_DCCP_ACKVEC */
+extern int dccp_ackvec_parsed_add(struct list_head *head,
+				  u8 *vec, u8 len, u8 nonce);
+extern void dccp_ackvec_parsed_cleanup(struct list_head *parsed_chunks);
 #endif /* _ACKVEC_H */

net/dccp/ccid.c

@@ -13,6 +13,13 @@
 
 #include "ccid.h"
 
+static u8 builtin_ccids[] = {
+	DCCPC_CCID2,		/* CCID2 is supported by default */
+#if defined(CONFIG_IP_DCCP_CCID3) || defined(CONFIG_IP_DCCP_CCID3_MODULE)
+	DCCPC_CCID3,
+#endif
+};
+
 static struct ccid_operations *ccids[CCID_MAX];
 #if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT)
 static atomic_t ccids_lockct = ATOMIC_INIT(0);
@@ -86,6 +93,47 @@ static void ccid_kmem_cache_destroy(struct kmem_cache *slab)
 	}
 }
 
+/* check that up to @array_len members in @ccid_array are supported */
+bool ccid_support_check(u8 const *ccid_array, u8 array_len)
+{
+	u8 i, j, found;
+
+	for (i = 0, found = 0; i < array_len; i++, found = 0) {
+		for (j = 0; !found && j < ARRAY_SIZE(builtin_ccids); j++)
+			found = (ccid_array[i] == builtin_ccids[j]);
+		if (!found)
+			return false;
+	}
+	return true;
+}
+
+/**
+ * ccid_get_builtin_ccids  -  Provide copy of `builtin' CCID array
+ * @ccid_array: pointer to copy into
+ * @array_len: value to return length into
+ * This function allocates memory - caller must see that it is freed after use.
+ */
+int ccid_get_builtin_ccids(u8 **ccid_array, u8 *array_len)
+{
+	*ccid_array = kmemdup(builtin_ccids, sizeof(builtin_ccids), gfp_any());
+	if (*ccid_array == NULL)
+		return -ENOBUFS;
+	*array_len = ARRAY_SIZE(builtin_ccids);
+	return 0;
+}
+
+int ccid_getsockopt_builtin_ccids(struct sock *sk, int len,
+				    char __user *optval, int __user *optlen)
+{
+	if (len < sizeof(builtin_ccids))
+		return -EINVAL;
+
+	if (put_user(sizeof(builtin_ccids), optlen) ||
+	    copy_to_user(optval, builtin_ccids, sizeof(builtin_ccids)))
+		return -EFAULT;
+	return 0;
+}
+
 int ccid_register(struct ccid_operations *ccid_ops)
 {
 	int err = -ENOBUFS;
@@ -148,22 +196,41 @@ int ccid_unregister(struct ccid_operations *ccid_ops)
 
 EXPORT_SYMBOL_GPL(ccid_unregister);
 
+/**
+ * ccid_request_module  -  Pre-load CCID module for later use
+ * This should be called only from process context (e.g. during connection
+ * setup) and is necessary for later calls to ccid_new (typically in software
+ * interrupt), so that it has the modules available when they are needed.
+ */
+static int ccid_request_module(u8 id)
+{
+	if (!in_atomic()) {
+		ccids_read_lock();
+		if (ccids[id] == NULL) {
+			ccids_read_unlock();
+			return request_module("net-dccp-ccid-%d", id);
+		}
+		ccids_read_unlock();
+	}
+	return 0;
+}
+
+int ccid_request_modules(u8 const *ccid_array, u8 array_len)
+{
+#ifdef CONFIG_KMOD
+	while (array_len--)
+		if (ccid_request_module(ccid_array[array_len]))
+			return -1;
+#endif
+	return 0;
+}
+
 struct ccid *ccid_new(unsigned char id, struct sock *sk, int rx, gfp_t gfp)
 {
 	struct ccid_operations *ccid_ops;
 	struct ccid *ccid = NULL;
 
 	ccids_read_lock();
-#ifdef CONFIG_KMOD
-	if (ccids[id] == NULL) {
-		/* We only try to load if in process context */
-		ccids_read_unlock();
-		if (gfp & GFP_ATOMIC)
-			goto out;
-		request_module("net-dccp-ccid-%d", id);
-		ccids_read_lock();
-	}
-#endif
 	ccid_ops = ccids[id];
 	if (ccid_ops == NULL)
 		goto out_unlock;
@@ -205,20 +272,6 @@ struct ccid *ccid_new(unsigned char id, struct sock *sk, int rx, gfp_t gfp)
 
 EXPORT_SYMBOL_GPL(ccid_new);
 
-struct ccid *ccid_hc_rx_new(unsigned char id, struct sock *sk, gfp_t gfp)
-{
-	return ccid_new(id, sk, 1, gfp);
-}
-
-EXPORT_SYMBOL_GPL(ccid_hc_rx_new);
-
-struct ccid *ccid_hc_tx_new(unsigned char id,struct sock *sk,  gfp_t gfp)
-{
-	return ccid_new(id, sk, 0, gfp);
-}
-
-EXPORT_SYMBOL_GPL(ccid_hc_tx_new);
-
 static void ccid_delete(struct ccid *ccid, struct sock *sk, int rx)
 {
 	struct ccid_operations *ccid_ops;

net/dccp/ccid.h

@@ -60,22 +60,18 @@ struct ccid_operations {
 	void		(*ccid_hc_tx_exit)(struct sock *sk);
 	void		(*ccid_hc_rx_packet_recv)(struct sock *sk,
 						  struct sk_buff *skb);
-	int		(*ccid_hc_rx_parse_options)(struct sock *sk,
-						    unsigned char option,
-						    unsigned char len, u16 idx,
-						    unsigned char* value);
+	int		(*ccid_hc_rx_parse_options)(struct sock *sk, u8 pkt,
+						    u8 opt, u8 *val, u8 len);
 	int		(*ccid_hc_rx_insert_options)(struct sock *sk,
 						     struct sk_buff *skb);
 	void		(*ccid_hc_tx_packet_recv)(struct sock *sk,
 						  struct sk_buff *skb);
-	int		(*ccid_hc_tx_parse_options)(struct sock *sk,
-						    unsigned char option,
-						    unsigned char len, u16 idx,
-						    unsigned char* value);
+	int		(*ccid_hc_tx_parse_options)(struct sock *sk, u8 pkt,
+						    u8 opt, u8 *val, u8 len);
 	int		(*ccid_hc_tx_send_packet)(struct sock *sk,
 						  struct sk_buff *skb);
 	void		(*ccid_hc_tx_packet_sent)(struct sock *sk,
-						  int more, unsigned int len);
+						  unsigned int len);
 	void		(*ccid_hc_rx_get_info)(struct sock *sk,
 					       struct tcp_info *info);
 	void		(*ccid_hc_tx_get_info)(struct sock *sk,
@@ -103,31 +99,78 @@ static inline void *ccid_priv(const struct ccid *ccid)
 	return (void *)ccid->ccid_priv;
 }
 
+extern bool ccid_support_check(u8 const *ccid_array, u8 array_len);
+extern int  ccid_get_builtin_ccids(u8 **ccid_array, u8 *array_len);
+extern int  ccid_getsockopt_builtin_ccids(struct sock *sk, int len,
+					  char __user *, int __user *);
+
+extern int    ccid_request_modules(u8 const *ccid_array, u8 array_len);
 extern struct ccid *ccid_new(unsigned char id, struct sock *sk, int rx,
 			     gfp_t gfp);
 
-extern struct ccid *ccid_hc_rx_new(unsigned char id, struct sock *sk,
-				   gfp_t gfp);
-extern struct ccid *ccid_hc_tx_new(unsigned char id, struct sock *sk,
-				   gfp_t gfp);
+static inline int ccid_get_current_rx_ccid(struct dccp_sock *dp)
+{
+	struct ccid *ccid = dp->dccps_hc_rx_ccid;
+
+	if (ccid == NULL || ccid->ccid_ops == NULL)
+		return -1;
+	return ccid->ccid_ops->ccid_id;
+}
+
+static inline int ccid_get_current_tx_ccid(struct dccp_sock *dp)
+{
+	struct ccid *ccid = dp->dccps_hc_tx_ccid;
+
+	if (ccid == NULL || ccid->ccid_ops == NULL)
+		return -1;
+	return ccid->ccid_ops->ccid_id;
+}
 
 extern void ccid_hc_rx_delete(struct ccid *ccid, struct sock *sk);
 extern void ccid_hc_tx_delete(struct ccid *ccid, struct sock *sk);
 
+/*
+ * Congestion control of queued data packets via CCID decision.
+ *
+ * The TX CCID performs its congestion-control by indicating whether and when a
+ * queued packet may be sent, using the return code of ccid_hc_tx_send_packet().
+ * The following modes are supported via the symbolic constants below:
+ * - timer-based pacing    (CCID returns a delay value in milliseconds);
+ * - autonomous dequeueing (CCID internally schedules dccps_xmitlet).
+ */
+
+enum ccid_dequeueing_decision {
+	CCID_PACKET_SEND_AT_ONCE =	 0x00000,  /* "green light": no delay */
+	CCID_PACKET_DELAY_MAX =		 0x0FFFF,  /* maximum delay in msecs  */
+	CCID_PACKET_DELAY =		 0x10000,  /* CCID msec-delay mode */
+	CCID_PACKET_WILL_DEQUEUE_LATER = 0x20000,  /* CCID autonomous mode */
+	CCID_PACKET_ERR =		 0xF0000,  /* error condition */
+};
+
+static inline int ccid_packet_dequeue_eval(const int return_code)
+{
+	if (return_code < 0)
+		return CCID_PACKET_ERR;
+	if (return_code == 0)
+		return CCID_PACKET_SEND_AT_ONCE;
+	if (return_code <= CCID_PACKET_DELAY_MAX)
+		return CCID_PACKET_DELAY;
+	return return_code;
+}
+
 static inline int ccid_hc_tx_send_packet(struct ccid *ccid, struct sock *sk,
 					 struct sk_buff *skb)
 {
-	int rc = 0;
 	if (ccid->ccid_ops->ccid_hc_tx_send_packet != NULL)
-		rc = ccid->ccid_ops->ccid_hc_tx_send_packet(sk, skb);
-	return rc;
+		return ccid->ccid_ops->ccid_hc_tx_send_packet(sk, skb);
+	return CCID_PACKET_SEND_AT_ONCE;
 }
 
 static inline void ccid_hc_tx_packet_sent(struct ccid *ccid, struct sock *sk,
-					  int more, unsigned int len)
+					  unsigned int len)
 {
 	if (ccid->ccid_ops->ccid_hc_tx_packet_sent != NULL)
-		ccid->ccid_ops->ccid_hc_tx_packet_sent(sk, more, len);
+		ccid->ccid_ops->ccid_hc_tx_packet_sent(sk, len);
 }
 
 static inline void ccid_hc_rx_packet_recv(struct ccid *ccid, struct sock *sk,
@@ -144,27 +187,31 @@ static inline void ccid_hc_tx_packet_recv(struct ccid *ccid, struct sock *sk,
 		ccid->ccid_ops->ccid_hc_tx_packet_recv(sk, skb);
 }
 
+/**
+ * ccid_hc_tx_parse_options  -  Parse CCID-specific options sent by the receiver
+ * @pkt: type of packet that @opt appears on (RFC 4340, 5.1)
+ * @opt: the CCID-specific option type (RFC 4340, 5.8 and 10.3)
+ * @val: value of @opt
+ * @len: length of @val in bytes
+ */
 static inline int ccid_hc_tx_parse_options(struct ccid *ccid, struct sock *sk,
-					   unsigned char option,
-					   unsigned char len, u16 idx,
-					   unsigned char* value)
+					   u8 pkt, u8 opt, u8 *val, u8 len)
 {
-	int rc = 0;
-	if (ccid->ccid_ops->ccid_hc_tx_parse_options != NULL)
-		rc = ccid->ccid_ops->ccid_hc_tx_parse_options(sk, option, len, idx,
-						    value);
-	return rc;
+	if (ccid->ccid_ops->ccid_hc_tx_parse_options == NULL)
+		return 0;
+	return ccid->ccid_ops->ccid_hc_tx_parse_options(sk, pkt, opt, val, len);
 }
 
+/**
+ * ccid_hc_rx_parse_options  -  Parse CCID-specific options sent by the sender
+ * Arguments are analogous to ccid_hc_tx_parse_options()
+ */
 static inline int ccid_hc_rx_parse_options(struct ccid *ccid, struct sock *sk,
-					   unsigned char option,
-					   unsigned char len, u16 idx,
-					   unsigned char* value)
+					   u8 pkt, u8 opt, u8 *val, u8 len)
 {
-	int rc = 0;
-	if (ccid->ccid_ops->ccid_hc_rx_parse_options != NULL)
-		rc = ccid->ccid_ops->ccid_hc_rx_parse_options(sk, option, len, idx, value);
-	return rc;
+	if (ccid->ccid_ops->ccid_hc_rx_parse_options == NULL)
+		return 0;
+	return ccid->ccid_ops->ccid_hc_rx_parse_options(sk, pkt, opt, val, len);
 }
 
 static inline int ccid_hc_rx_insert_options(struct ccid *ccid, struct sock *sk,

net/dccp/ccids/Kconfig

 menu "DCCP CCIDs Configuration (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
 
 config IP_DCCP_CCID2
-	tristate "CCID2 (TCP-Like) (EXPERIMENTAL)"
+	tristate "CCID2 (TCP-Like)"
 	def_tristate IP_DCCP
-	select IP_DCCP_ACKVEC
 	---help---
 	  CCID 2, TCP-like Congestion Control, denotes Additive Increase,
 	  Multiplicative Decrease (AIMD) congestion control with behavior
@@ -36,7 +34,7 @@ config IP_DCCP_CCID2_DEBUG
 	    If in doubt, say N.
 
 config IP_DCCP_CCID3
-	tristate "CCID3 (TCP-Friendly) (EXPERIMENTAL)"
+	tristate "CCID3 (TCP-Friendly)"
 	def_tristate IP_DCCP
 	select IP_DCCP_TFRC_LIB
 	---help---
@@ -64,9 +62,9 @@ config IP_DCCP_CCID3
 
 	  If in doubt, say M.
 
+if IP_DCCP_CCID3
 config IP_DCCP_CCID3_DEBUG
 	  bool "CCID3 debugging messages"
-	  depends on IP_DCCP_CCID3
 	  ---help---
 	    Enable CCID3-specific debugging messages.
 
@@ -76,10 +74,29 @@ config IP_DCCP_CCID3_DEBUG
 
 	    If in doubt, say N.
 
+choice
+	  prompt "Select method for measuring the packet size s"
+	  default IP_DCCP_CCID3_MEASURE_S_AS_MPS
+
+config IP_DCCP_CCID3_MEASURE_S_AS_MPS
+	  bool "Always use MPS in place of s"
+	  ---help---
+	    This use is recommended as it is consistent with the initialisation
+	    of X and suggested when s varies (rfc3448bis, (1) in section 4.1).
+config IP_DCCP_CCID3_MEASURE_S_AS_AVG
+	  bool "Use moving average"
+	  ---help---
+	    An alternative way of tracking s, also supported by rfc3448bis.
+	    This used to be the default for CCID-3 in previous kernels.
+config IP_DCCP_CCID3_MEASURE_S_AS_MAX
+	  bool "Track the maximum payload length"
+	  ---help---
+	    An experimental method based on tracking the maximum packet size.
+endchoice
+
 config IP_DCCP_CCID3_RTO
 	  int "Use higher bound for nofeedback timer"
 	  default 100
-	  depends on IP_DCCP_CCID3 && EXPERIMENTAL
 	  ---help---
 	    Use higher lower bound for nofeedback timer expiration.
 
@@ -106,6 +123,7 @@ config IP_DCCP_CCID3_RTO
 	    The purpose of the nofeedback timer is to slow DCCP down when there
 	    is serious network congestion: experimenting with larger values should
 	    therefore not be performed on WANs.
+endif	# IP_DCCP_CCID3
 
 config IP_DCCP_TFRC_LIB
 	tristate

net/dccp/ccids/ccid2.h

@@ -42,34 +42,49 @@ struct ccid2_seq {
 
 /** struct ccid2_hc_tx_sock - CCID2 TX half connection
  *
- * @ccid2hctx_{cwnd,ssthresh,pipe}: as per RFC 4341, section 5
- * @ccid2hctx_packets_acked - Ack counter for deriving cwnd growth (RFC 3465)
- * @ccid2hctx_lastrtt -time RTT was last measured
- * @ccid2hctx_rpseq - last consecutive seqno
- * @ccid2hctx_rpdupack - dupacks since rpseq
-*/
+ * @{cwnd,ssthresh,pipe}: as per RFC 4341, section 5
+ * @packets_acked: Ack counter for deriving cwnd growth (RFC 3465)
+ * @srtt: smoothed RTT estimate, scaled by 2^3
+ * @mdev: smoothed RTT variation, scaled by 2^2
+ * @mdev_max: maximum of @mdev during one flight
+ * @rttvar: moving average/maximum of @mdev_max
+ * @rto: RTO value deriving from SRTT and RTTVAR (RFC 2988)
+ * @rtt_seq: to decay RTTVAR at most once per flight
+ * @rpseq: last consecutive seqno
+ * @rpdupack: dupacks since rpseq
+ * @av_chunks: list of Ack Vectors received on current skb
+ */
 struct ccid2_hc_tx_sock {
-	u32			ccid2hctx_cwnd;
-	u32			ccid2hctx_ssthresh;
-	u32			ccid2hctx_pipe;
-	u32			ccid2hctx_packets_acked;
-	struct ccid2_seq	*ccid2hctx_seqbuf[CCID2_SEQBUF_MAX];
-	int			ccid2hctx_seqbufc;
-	struct ccid2_seq	*ccid2hctx_seqh;
-	struct ccid2_seq	*ccid2hctx_seqt;
-	long			ccid2hctx_rto;
-	long			ccid2hctx_srtt;
-	long			ccid2hctx_rttvar;
-	unsigned long		ccid2hctx_lastrtt;
-	struct timer_list	ccid2hctx_rtotimer;
-	u64			ccid2hctx_rpseq;
-	int			ccid2hctx_rpdupack;
-	unsigned long		ccid2hctx_last_cong;
-	u64			ccid2hctx_high_ack;
+	u32			cwnd;
+	u32			ssthresh;
+	u32			pipe;
+	u32			packets_acked;
+	struct ccid2_seq	*seqbuf[CCID2_SEQBUF_MAX];
+	int			seqbufc;
+	struct ccid2_seq	*seqh;
+	struct ccid2_seq	*seqt;
+	/* RTT measurement: variables/principles are the same as in TCP */
+	u32			srtt,
+				mdev,
+				mdev_max,
+				rttvar,
+				rto;
+	u64			rtt_seq:48;
+	struct timer_list	rtotimer;
+	u64			rpseq;
+	int			rpdupack;
+	unsigned long		last_cong;
+	u64			high_ack;
+	struct list_head	av_chunks;
 };
 
+static inline bool ccid2_cwnd_network_limited(struct ccid2_hc_tx_sock *hctx)
+{
+	return (hctx->pipe >= hctx->cwnd);
+}
+
 struct ccid2_hc_rx_sock {
-	int	ccid2hcrx_data;
+	int			data;
 };
 
 static inline struct ccid2_hc_tx_sock *ccid2_hc_tx_sk(const struct sock *sk)

net/dccp/ccids/ccid3.h

@@ -47,11 +47,22 @@
 /* Two seconds as per RFC 3448 4.2 */
 #define TFRC_INITIAL_TIMEOUT	   (2 * USEC_PER_SEC)
 
-/* In usecs - half the scheduling granularity as per RFC3448 4.6 */
-#define TFRC_OPSYS_HALF_TIME_GRAN  (USEC_PER_SEC / (2 * HZ))
+/* Maximum backoff interval t_mbi (RFC 3448, 4.3) */
+#define TFRC_T_MBI		   (64 * USEC_PER_SEC)
 
-/* Parameter t_mbi from [RFC 3448, 4.3]: backoff interval in seconds */
-#define TFRC_T_MBI		   64
+/*
+ * The t_delta parameter (RFC 3448, 4.6): delays of less than %USEC_PER_MSEC are
+ * rounded down to 0, since sk_reset_timer() here uses millisecond granularity.
+ * Hence we can use a constant t_delta = %USEC_PER_MSEC when HZ >= 500. A coarse
+ * resolution of HZ < 500 means that the error is below one timer tick (t_gran)
+ * when using the constant t_delta  =  t_gran / 2  =  %USEC_PER_SEC / (2 * HZ).
+ */
+#if (HZ >= 500)
+# define TFRC_T_DELTA		   USEC_PER_MSEC
+#else
+# define TFRC_T_DELTA		   (USEC_PER_SEC / (2 * HZ))
+#warning Coarse CONFIG_HZ resolution -- higher value recommended for TFRC.
+#endif
 
 enum ccid3_options {
 	TFRC_OPT_LOSS_EVENT_RATE = 192,
@@ -59,62 +70,43 @@ enum ccid3_options {
 	TFRC_OPT_RECEIVE_RATE	 = 194,
 };
 
-struct ccid3_options_received {
-	u64 ccid3or_seqno:48,
-	    ccid3or_loss_intervals_idx:16;
-	u16 ccid3or_loss_intervals_len;
-	u32 ccid3or_loss_event_rate;
-	u32 ccid3or_receive_rate;
-};
-
-/* TFRC sender states */
-enum ccid3_hc_tx_states {
-	TFRC_SSTATE_NO_SENT = 1,
-	TFRC_SSTATE_NO_FBACK,
-	TFRC_SSTATE_FBACK,
-	TFRC_SSTATE_TERM,
-};
-
 /** struct ccid3_hc_tx_sock - CCID3 sender half-connection socket
  *
- * @ccid3hctx_x - Current sending rate in 64 * bytes per second
- * @ccid3hctx_x_recv - Receive rate    in 64 * bytes per second
- * @ccid3hctx_x_calc - Calculated rate in bytes per second
- * @ccid3hctx_rtt - Estimate of current round trip time in usecs
- * @ccid3hctx_p - Current loss event rate (0-1) scaled by 1000000
- * @ccid3hctx_s - Packet size in bytes
- * @ccid3hctx_t_rto - Nofeedback Timer setting in usecs
- * @ccid3hctx_t_ipi - Interpacket (send) interval (RFC 3448, 4.6) in usecs
- * @ccid3hctx_state - Sender state, one of %ccid3_hc_tx_states
- * @ccid3hctx_last_win_count - Last window counter sent
- * @ccid3hctx_t_last_win_count - Timestamp of earliest packet
- *				 with last_win_count value sent
- * @ccid3hctx_no_feedback_timer - Handle to no feedback timer
- * @ccid3hctx_t_ld - Time last doubled during slow start
- * @ccid3hctx_t_nom - Nominal send time of next packet
- * @ccid3hctx_delta - Send timer delta (RFC 3448, 4.6) in usecs
- * @ccid3hctx_hist - Packet history
- * @ccid3hctx_options_received - Parsed set of retrieved options
+ * @x - Current sending rate in 64 * bytes per second
+ * @x_recv - Receive rate    in 64 * bytes per second
+ * @x_calc - Calculated rate in bytes per second
+ * @rtt - Estimate of current round trip time in usecs
+ * @r_sqmean - Estimate of long-term RTT (RFC 3448, 4.5)
+ * @p - Current loss event rate (0-1) scaled by 1000000
+ * @s - Packet size in bytes
+ * @t_rto - Nofeedback Timer setting in usecs
+ * @t_ipi - Interpacket (send) interval (RFC 3448, 4.6) in usecs
+ * @feedback - Whether feedback has been received or not
+ * @last_win_count - Last window counter sent
+ * @t_last_win_count - Timestamp of earliest packet with
+ *                     last_win_count value sent
+ * @no_feedback_timer - Handle to no feedback timer
+ * @t_ld - Time last doubled during slow start
+ * @t_nom - Nominal send time of next packet
+ * @hist - Packet history
  */
 struct ccid3_hc_tx_sock {
-	struct tfrc_tx_info		ccid3hctx_tfrc;
-#define ccid3hctx_x			ccid3hctx_tfrc.tfrctx_x
-#define ccid3hctx_x_recv		ccid3hctx_tfrc.tfrctx_x_recv
-#define ccid3hctx_x_calc		ccid3hctx_tfrc.tfrctx_x_calc
-#define ccid3hctx_rtt			ccid3hctx_tfrc.tfrctx_rtt
-#define ccid3hctx_p			ccid3hctx_tfrc.tfrctx_p
-#define ccid3hctx_t_rto			ccid3hctx_tfrc.tfrctx_rto
-#define ccid3hctx_t_ipi			ccid3hctx_tfrc.tfrctx_ipi
-	u16				ccid3hctx_s;
-	enum ccid3_hc_tx_states		ccid3hctx_state:8;
-	u8				ccid3hctx_last_win_count;
-	ktime_t				ccid3hctx_t_last_win_count;
-	struct timer_list		ccid3hctx_no_feedback_timer;
-	ktime_t				ccid3hctx_t_ld;
-	ktime_t				ccid3hctx_t_nom;
-	u32				ccid3hctx_delta;
-	struct tfrc_tx_hist_entry	*ccid3hctx_hist;
-	struct ccid3_options_received	ccid3hctx_options_received;
+	u64				x;
+	u64				x_recv;
+	u32				x_calc;
+	u32				rtt;
+	u16				r_sqmean;
+	u32				p;
+	u32				t_rto;
+	u32				t_ipi;
+	u16				s;
+	bool				feedback:1;
+	u8				last_win_count;
+	ktime_t				t_last_win_count;
+	struct timer_list		no_feedback_timer;
+	ktime_t				t_ld;
+	ktime_t				t_nom;
+	struct tfrc_tx_hist_entry	*hist;
 };
 
 static inline struct ccid3_hc_tx_sock *ccid3_hc_tx_sk(const struct sock *sk)
@@ -124,41 +116,32 @@ static inline struct ccid3_hc_tx_sock *ccid3_hc_tx_sk(const struct sock *sk)
     return hctx;
 }
 
-/* TFRC receiver states */
-enum ccid3_hc_rx_states {
-	TFRC_RSTATE_NO_DATA = 1,
-	TFRC_RSTATE_DATA,
-	TFRC_RSTATE_TERM    = 127,
+
+enum ccid3_fback_type {
+	CCID3_FBACK_NONE = 0,
+	CCID3_FBACK_INITIAL,
+	CCID3_FBACK_PERIODIC,
+	CCID3_FBACK_PARAM_CHANGE
 };
 
 /** struct ccid3_hc_rx_sock - CCID3 receiver half-connection socket
  *
- *  @ccid3hcrx_x_recv  -  Receiver estimate of send rate (RFC 3448 4.3)
- *  @ccid3hcrx_rtt  -  Receiver estimate of rtt (non-standard)
- *  @ccid3hcrx_p  -  Current loss event rate (RFC 3448 5.4)
- *  @ccid3hcrx_last_counter  -  Tracks window counter (RFC 4342, 8.1)
- *  @ccid3hcrx_state  -  Receiver state, one of %ccid3_hc_rx_states
- *  @ccid3hcrx_bytes_recv  -  Total sum of DCCP payload bytes
- *  @ccid3hcrx_x_recv  -  Receiver estimate of send rate (RFC 3448, sec. 4.3)
- *  @ccid3hcrx_rtt  -  Receiver estimate of RTT
- *  @ccid3hcrx_tstamp_last_feedback  -  Time at which last feedback was sent
- *  @ccid3hcrx_tstamp_last_ack  -  Time at which last feedback was sent
- *  @ccid3hcrx_hist  -  Packet history (loss detection + RTT sampling)
- *  @ccid3hcrx_li_hist  -  Loss Interval database
- *  @ccid3hcrx_s  -  Received packet size in bytes
- *  @ccid3hcrx_pinv  -  Inverse of Loss Event Rate (RFC 4342, sec. 8.5)
+ *  @last_counter  -  Tracks window counter (RFC 4342, 8.1)
+ *  @feedback  -  The type of the feedback last sent
+ *  @x_recv  -  Receiver estimate of send rate (RFC 3448, sec. 4.3)
+ *  @tstamp_last_feedback  -  Time at which last feedback was sent
+ *  @hist  -  Packet history (loss detection + RTT sampling)
+ *  @li_hist  -  Loss Interval database
+ *  @p_inverse  -  Inverse of Loss Event Rate (RFC 4342, sec. 8.5)
  */
 struct ccid3_hc_rx_sock {
-	u8				ccid3hcrx_last_counter:4;
-	enum ccid3_hc_rx_states		ccid3hcrx_state:8;
-	u32				ccid3hcrx_bytes_recv;
-	u32				ccid3hcrx_x_recv;
-	u32				ccid3hcrx_rtt;
-	ktime_t				ccid3hcrx_tstamp_last_feedback;
-	struct tfrc_rx_hist		ccid3hcrx_hist;
-	struct tfrc_loss_hist		ccid3hcrx_li_hist;
-	u16				ccid3hcrx_s;
-#define ccid3hcrx_pinv			ccid3hcrx_li_hist.i_mean
+	u8				last_counter:4;
+	enum ccid3_fback_type		feedback:4;
+	u32				x_recv;
+	ktime_t				tstamp_last_feedback;
+	struct tfrc_rx_hist		hist;
+	struct tfrc_loss_hist		li_hist;
+#define p_inverse			li_hist.i_mean
 };
 
 static inline struct ccid3_hc_rx_sock *ccid3_hc_rx_sk(const struct sock *sk)

net/dccp/ccids/lib/loss_interval.c

@@ -86,21 +86,26 @@ static void tfrc_lh_calc_i_mean(struct tfrc_loss_hist *lh)
 
 /**
  * tfrc_lh_update_i_mean  -  Update the `open' loss interval I_0
- * For recomputing p: returns `true' if p > p_prev  <=>  1/p < 1/p_prev
+ * This updates I_mean as the sequence numbers increase. As a consequence, the
+ * open loss interval I_0 increases, hence p = W_tot/max(I_tot0, I_tot1)
+ * decreases, and thus there is no need to send renewed feedback.
  */
-u8 tfrc_lh_update_i_mean(struct tfrc_loss_hist *lh, struct sk_buff *skb)
+void tfrc_lh_update_i_mean(struct tfrc_loss_hist *lh, struct sk_buff *skb)
 {
 	struct tfrc_loss_interval *cur = tfrc_lh_peek(lh);
-	u32 old_i_mean = lh->i_mean;
 	s64 len;
 
 	if (cur == NULL)			/* not initialised */
-		return 0;
+		return;
+
+	/* FIXME: should probably also count non-data packets (RFC 4342, 6.1) */
+	if (!dccp_data_packet(skb))
+		return;
 
 	len = dccp_delta_seqno(cur->li_seqno, DCCP_SKB_CB(skb)->dccpd_seq) + 1;
 
 	if (len - (s64)cur->li_length <= 0)	/* duplicate or reordered */
-		return 0;
+		return;
 
 	if (SUB16(dccp_hdr(skb)->dccph_ccval, cur->li_ccval) > 4)
 		/*
@@ -114,14 +119,11 @@ u8 tfrc_lh_update_i_mean(struct tfrc_loss_hist *lh, struct sk_buff *skb)
 		cur->li_is_closed = 1;
 
 	if (tfrc_lh_length(lh) == 1)		/* due to RFC 3448, 6.3.1 */
-		return 0;
+		return;
 
 	cur->li_length = len;
 	tfrc_lh_calc_i_mean(lh);
-
-	return (lh->i_mean < old_i_mean);
 }
-EXPORT_SYMBOL_GPL(tfrc_lh_update_i_mean);
 
 /* Determine if `new_loss' does begin a new loss interval [RFC 4342, 10.2] */
 static inline u8 tfrc_lh_is_new_loss(struct tfrc_loss_interval *cur,
@@ -138,18 +140,18 @@ static inline u8 tfrc_lh_is_new_loss(struct tfrc_loss_interval *cur,
  * @sk:		   Used by @calc_first_li in caller-specific way (subtyping)
  * Updates I_mean and returns 1 if a new interval has in fact been added to @lh.
  */
-int tfrc_lh_interval_add(struct tfrc_loss_hist *lh, struct tfrc_rx_hist *rh,
-			 u32 (*calc_first_li)(struct sock *), struct sock *sk)
+bool tfrc_lh_interval_add(struct tfrc_loss_hist *lh, struct tfrc_rx_hist *rh,
+			  u32 (*calc_first_li)(struct sock *), struct sock *sk)
 {
 	struct tfrc_loss_interval *cur = tfrc_lh_peek(lh), *new;
 
 	if (cur != NULL && !tfrc_lh_is_new_loss(cur, tfrc_rx_hist_loss_prev(rh)))
-		return 0;
+		return false;
 
 	new = tfrc_lh_demand_next(lh);
 	if (unlikely(new == NULL)) {
 		DCCP_CRIT("Cannot allocate/add loss record.");
-		return 0;
+		return false;
 	}
 
 	new->li_seqno	  = tfrc_rx_hist_loss_prev(rh)->tfrchrx_seqno;
@@ -167,7 +169,7 @@ int tfrc_lh_interval_add(struct tfrc_loss_hist *lh, struct tfrc_rx_hist *rh,
 
 		tfrc_lh_calc_i_mean(lh);
 	}
-	return 1;
+	return true;
 }
 EXPORT_SYMBOL_GPL(tfrc_lh_interval_add);
 

net/dccp/ccids/lib/loss_interval.h

@@ -67,9 +67,9 @@ static inline u8 tfrc_lh_length(struct tfrc_loss_hist *lh)
 
 struct tfrc_rx_hist;
 
-extern int  tfrc_lh_interval_add(struct tfrc_loss_hist *, struct tfrc_rx_hist *,
+extern bool tfrc_lh_interval_add(struct tfrc_loss_hist *, struct tfrc_rx_hist *,
 				 u32 (*first_li)(struct sock *), struct sock *);
-extern u8   tfrc_lh_update_i_mean(struct tfrc_loss_hist *lh, struct sk_buff *);
+extern void tfrc_lh_update_i_mean(struct tfrc_loss_hist *lh, struct sk_buff *);
 extern void tfrc_lh_cleanup(struct tfrc_loss_hist *lh);
 
 #endif /* _DCCP_LI_HIST_ */

net/dccp/ccids/lib/packet_history.c

@@ -40,18 +40,6 @@
 #include "packet_history.h"
 #include "../../dccp.h"
 
-/**
- *  tfrc_tx_hist_entry  -  Simple singly-linked TX history list
- *  @next:  next oldest entry (LIFO order)
- *  @seqno: sequence number of this entry
- *  @stamp: send time of packet with sequence number @seqno
- */
-struct tfrc_tx_hist_entry {
-	struct tfrc_tx_hist_entry *next;
-	u64			  seqno;
-	ktime_t			  stamp;
-};
-
 /*
  * Transmitter History Routines
  */
@@ -73,15 +61,6 @@ void tfrc_tx_packet_history_exit(void)
 	}
 }
 
-static struct tfrc_tx_hist_entry *
-	tfrc_tx_hist_find_entry(struct tfrc_tx_hist_entry *head, u64 seqno)
-{
-	while (head != NULL && head->seqno != seqno)
-		head = head->next;
-
-	return head;
-}
-
 int tfrc_tx_hist_add(struct tfrc_tx_hist_entry **headp, u64 seqno)
 {
 	struct tfrc_tx_hist_entry *entry = kmem_cache_alloc(tfrc_tx_hist_slab, gfp_any());
@@ -111,25 +90,6 @@ void tfrc_tx_hist_purge(struct tfrc_tx_hist_entry **headp)
 }
 EXPORT_SYMBOL_GPL(tfrc_tx_hist_purge);
 
-u32 tfrc_tx_hist_rtt(struct tfrc_tx_hist_entry *head, const u64 seqno,
-		     const ktime_t now)
-{
-	u32 rtt = 0;
-	struct tfrc_tx_hist_entry *packet = tfrc_tx_hist_find_entry(head, seqno);
-
-	if (packet != NULL) {
-		rtt = ktime_us_delta(now, packet->stamp);
-		/*
-		 * Garbage-collect older (irrelevant) entries:
-		 */
-		tfrc_tx_hist_purge(&packet->next);
-	}
-
-	return rtt;
-}
-EXPORT_SYMBOL_GPL(tfrc_tx_hist_rtt);
-
-
 /*
  * 	Receiver History Routines
  */
@@ -191,14 +151,31 @@ int tfrc_rx_hist_duplicate(struct tfrc_rx_hist *h, struct sk_buff *skb)
 }
 EXPORT_SYMBOL_GPL(tfrc_rx_hist_duplicate);
 
+
+static void __tfrc_rx_hist_swap(struct tfrc_rx_hist *h, const u8 a, const u8 b)
+{
+	struct tfrc_rx_hist_entry *tmp = h->ring[a];
+
+	h->ring[a] = h->ring[b];
+	h->ring[b] = tmp;
+}
+
 static void tfrc_rx_hist_swap(struct tfrc_rx_hist *h, const u8 a, const u8 b)
 {
-	const u8 idx_a = tfrc_rx_hist_index(h, a),
-		 idx_b = tfrc_rx_hist_index(h, b);
-	struct tfrc_rx_hist_entry *tmp = h->ring[idx_a];
+	__tfrc_rx_hist_swap(h, tfrc_rx_hist_index(h, a),
+			       tfrc_rx_hist_index(h, b));
+}
 
-	h->ring[idx_a] = h->ring[idx_b];
-	h->ring[idx_b] = tmp;
+/**
+ * tfrc_rx_hist_resume_rtt_sampling  -  Prepare RX history for RTT sampling
+ * This is called after loss detection has finished, when the history entry
+ * with the index of `loss_count' holds the highest-received sequence number.
+ * RTT sampling requires this information at ring[0] (tfrc_rx_hist_sample_rtt).
+ */
+static inline void tfrc_rx_hist_resume_rtt_sampling(struct tfrc_rx_hist *h)
+{
+	__tfrc_rx_hist_swap(h, 0, tfrc_rx_hist_index(h, h->loss_count));
+	h->loss_count = h->loss_start = 0;
 }
 
 /*
@@ -215,10 +192,8 @@ static void __do_track_loss(struct tfrc_rx_hist *h, struct sk_buff *skb, u64 n1)
 	u64 s0 = tfrc_rx_hist_loss_prev(h)->tfrchrx_seqno,
 	    s1 = DCCP_SKB_CB(skb)->dccpd_seq;
 
-	if (!dccp_loss_free(s0, s1, n1)) {	/* gap between S0 and S1 */
+	if (!dccp_loss_free(s0, s1, n1))	/* gap between S0 and S1 */
 		h->loss_count = 1;
-		tfrc_rx_hist_entry_from_skb(tfrc_rx_hist_entry(h, 1), skb, n1);
-	}
 }
 
 static void __one_after_loss(struct tfrc_rx_hist *h, struct sk_buff *skb, u32 n2)
@@ -240,8 +215,7 @@ static void __one_after_loss(struct tfrc_rx_hist *h, struct sk_buff *skb, u32 n2
 
 		if (dccp_loss_free(s2, s1, n1)) {
 			/* hole is filled: S0, S2, and S1 are consecutive */
-			h->loss_count = 0;
-			h->loss_start = tfrc_rx_hist_index(h, 1);
+			tfrc_rx_hist_resume_rtt_sampling(h);
 		} else
 			/* gap between S2 and S1: just update loss_prev */
 			tfrc_rx_hist_entry_from_skb(tfrc_rx_hist_loss_prev(h), skb, n2);
@@ -294,8 +268,7 @@ static int __two_after_loss(struct tfrc_rx_hist *h, struct sk_buff *skb, u32 n3)
 
 			if (dccp_loss_free(s1, s2, n2)) {
 				/* entire hole filled by S0, S3, S1, S2 */
-				h->loss_start = tfrc_rx_hist_index(h, 2);
-				h->loss_count = 0;
+				tfrc_rx_hist_resume_rtt_sampling(h);
 			} else {
 				/* gap remains between S1 and S2 */
 				h->loss_start = tfrc_rx_hist_index(h, 1);
@@ -339,8 +312,7 @@ static void __three_after_loss(struct tfrc_rx_hist *h)
 
 		if (dccp_loss_free(s2, s3, n3)) {
 			/* no gap between S2 and S3: entire hole is filled */
-			h->loss_start = tfrc_rx_hist_index(h, 3);
-			h->loss_count = 0;
+			tfrc_rx_hist_resume_rtt_sampling(h);
 		} else {
 			/* gap between S2 and S3 */
 			h->loss_start = tfrc_rx_hist_index(h, 2);
@@ -354,13 +326,13 @@ static void __three_after_loss(struct tfrc_rx_hist *h)
 }
 
 /**
- *  tfrc_rx_handle_loss  -  Loss detection and further processing
- *  @h:		    The non-empty RX history object
- *  @lh:	    Loss Intervals database to update
- *  @skb:	    Currently received packet
- *  @ndp:	    The NDP count belonging to @skb
- *  @calc_first_li: Caller-dependent computation of first loss interval in @lh
- *  @sk:	    Used by @calc_first_li (see tfrc_lh_interval_add)
+ *  tfrc_rx_congestion_event  -  Loss detection and further processing
+ *  @h:		The non-empty RX history object
+ *  @lh:	Loss Intervals database to update
+ *  @skb:	Currently received packet
+ *  @ndp:	The NDP count belonging to @skb
+ *  @first_li:	Caller-dependent computation of first loss interval in @lh
+ *  @sk:	Used by @calc_first_li (see tfrc_lh_interval_add)
  *  Chooses action according to pending loss, updates LI database when a new
  *  loss was detected, and does required post-processing. Returns 1 when caller
  *  should send feedback, 0 otherwise.
@@ -368,15 +340,20 @@ static void __three_after_loss(struct tfrc_rx_hist *h)
  *  records accordingly, the caller should not perform any more RX history
  *  operations when loss_count is greater than 0 after calling this function.
  */
-int tfrc_rx_handle_loss(struct tfrc_rx_hist *h,
-			struct tfrc_loss_hist *lh,
-			struct sk_buff *skb, const u64 ndp,
-			u32 (*calc_first_li)(struct sock *), struct sock *sk)
+bool tfrc_rx_congestion_event(struct tfrc_rx_hist *h,
+			      struct tfrc_loss_hist *lh,
+			      struct sk_buff *skb, const u64 ndp,
+			      u32 (*first_li)(struct sock *), struct sock *sk)
 {
-	int is_new_loss = 0;
+	bool new_event = false;
+
+	if (tfrc_rx_hist_duplicate(h, skb))
+		return 0;
 
 	if (h->loss_count == 0) {
 		__do_track_loss(h, skb, ndp);
+		tfrc_rx_hist_sample_rtt(h, skb);
+		tfrc_rx_hist_add_packet(h, skb, ndp);
 	} else if (h->loss_count == 1) {
 		__one_after_loss(h, skb, ndp);
 	} else if (h->loss_count != 2) {
@@ -385,34 +362,57 @@ int tfrc_rx_handle_loss(struct tfrc_rx_hist *h,
 		/*
 		 * Update Loss Interval database and recycle RX records
 		 */
-		is_new_loss = tfrc_lh_interval_add(lh, h, calc_first_li, sk);
+		new_event = tfrc_lh_interval_add(lh, h, first_li, sk);
 		__three_after_loss(h);
 	}
-	return is_new_loss;
+
+	/*
+	 * Update moving-average of `s' and the sum of received payload bytes.
+	 */
+	if (dccp_data_packet(skb)) {
+		const u32 payload = skb->len - dccp_hdr(skb)->dccph_doff * 4;
+
+		h->packet_size = tfrc_ewma(h->packet_size, payload, 9);
+		h->bytes_recvd += payload;
+	}
+
+	/* RFC 3448, 6.1: update I_0, whose growth implies p <= p_prev */
+	if (!new_event)
+		tfrc_lh_update_i_mean(lh, skb);
+
+	return new_event;
 }
-EXPORT_SYMBOL_GPL(tfrc_rx_handle_loss);
+EXPORT_SYMBOL_GPL(tfrc_rx_congestion_event);
 
-int tfrc_rx_hist_alloc(struct tfrc_rx_hist *h)
+/* Compute the sending rate X_recv measured between feedback intervals */
+u32 tfrc_rx_hist_x_recv(struct tfrc_rx_hist *h, const u32 last_x_recv)
 {
-	int i;
+	u64 bytes = h->bytes_recvd, last_rtt = h->rtt_estimate;
+	s64 delta = ktime_to_us(net_timedelta(h->bytes_start));
 
-	for (i = 0; i <= TFRC_NDUPACK; i++) {
-		h->ring[i] = kmem_cache_alloc(tfrc_rx_hist_slab, GFP_ATOMIC);
-		if (h->ring[i] == NULL)
-			goto out_free;
-	}
+	WARN_ON(delta <= 0);
+	/*
+	 * Ensure that the sampling interval for X_recv is at least one RTT,
+	 * by extending the sampling interval backwards in time, over the last
+	 * R_(m-1) seconds, as per rfc3448bis-06, 6.2.
+	 * To reduce noise (e.g. when the RTT changes often), this is only
+	 * done when delta is smaller than RTT/2.
+	 */
+	if (last_x_recv > 0 && delta < last_rtt/2) {
+		tfrc_pr_debug("delta < RTT ==> %ld us < %u us\n",
+			      (long)delta, (unsigned)last_rtt);
 
-	h->loss_count = h->loss_start = 0;
-	return 0;
+		delta = (bytes ? delta : 0) + last_rtt;
+		bytes += div_u64((u64)last_x_recv * last_rtt, USEC_PER_SEC);
+	}
 
-out_free:
-	while (i-- != 0) {
-		kmem_cache_free(tfrc_rx_hist_slab, h->ring[i]);
-		h->ring[i] = NULL;
+	if (unlikely(bytes == 0)) {
+		DCCP_WARN("X_recv == 0, using old value of %u\n", last_x_recv);
+		return last_x_recv;
 	}
-	return -ENOBUFS;
+	return scaled_div32(bytes, delta);
 }
-EXPORT_SYMBOL_GPL(tfrc_rx_hist_alloc);
+EXPORT_SYMBOL_GPL(tfrc_rx_hist_x_recv);
 
 void tfrc_rx_hist_purge(struct tfrc_rx_hist *h)
 {
@@ -426,73 +426,81 @@ void tfrc_rx_hist_purge(struct tfrc_rx_hist *h)
 }
 EXPORT_SYMBOL_GPL(tfrc_rx_hist_purge);
 
-/**
- * tfrc_rx_hist_rtt_last_s - reference entry to compute RTT samples against
- */
-static inline struct tfrc_rx_hist_entry *
-			tfrc_rx_hist_rtt_last_s(const struct tfrc_rx_hist *h)
+static int tfrc_rx_hist_alloc(struct tfrc_rx_hist *h)
 {
-	return h->ring[0];
+	int i;
+
+	memset(h, 0, sizeof(*h));
+
+	for (i = 0; i <= TFRC_NDUPACK; i++) {
+		h->ring[i] = kmem_cache_alloc(tfrc_rx_hist_slab, GFP_ATOMIC);
+		if (h->ring[i] == NULL) {
+			tfrc_rx_hist_purge(h);
+			return -ENOBUFS;
+		}
+	}
+	return 0;
 }
 
-/**
- * tfrc_rx_hist_rtt_prev_s: previously suitable (wrt rtt_last_s) RTT-sampling entry
- */
-static inline struct tfrc_rx_hist_entry *
-			tfrc_rx_hist_rtt_prev_s(const struct tfrc_rx_hist *h)
+int tfrc_rx_hist_init(struct tfrc_rx_hist *h, struct sock *sk)
 {
-	return h->ring[h->rtt_sample_prev];
+	if (tfrc_rx_hist_alloc(h))
+		return -ENOBUFS;
+	/*
+	 * Initialise first entry with GSR to start loss detection as early as
+	 * possible. Code using this must not use any other fields. The entry
+	 * will be overwritten once the CCID updates its received packets.
+	 */
+	tfrc_rx_hist_loss_prev(h)->tfrchrx_seqno = dccp_sk(sk)->dccps_gsr;
+	return 0;
 }
+EXPORT_SYMBOL_GPL(tfrc_rx_hist_init);
 
 /**
  * tfrc_rx_hist_sample_rtt  -  Sample RTT from timestamp / CCVal
- * Based on ideas presented in RFC 4342, 8.1. Returns 0 if it was not able
- * to compute a sample with given data - calling function should check this.
+ * Based on ideas presented in RFC 4342, 8.1. This function expects that no loss
+ * is pending and uses the following history entries (via rtt_sample_prev):
+ * - h->ring[0]  contains the most recent history entry prior to @skb;
+ * - h->ring[1]  is an unused `dummy' entry when the current difference is 0;
  */
-u32 tfrc_rx_hist_sample_rtt(struct tfrc_rx_hist *h, const struct sk_buff *skb)
+void tfrc_rx_hist_sample_rtt(struct tfrc_rx_hist *h, const struct sk_buff *skb)
 {
-	u32 sample = 0,
-	    delta_v = SUB16(dccp_hdr(skb)->dccph_ccval,
-			    tfrc_rx_hist_rtt_last_s(h)->tfrchrx_ccval);
-
-	if (delta_v < 1 || delta_v > 4) {	/* unsuitable CCVal delta */
-		if (h->rtt_sample_prev == 2) {	/* previous candidate stored */
-			sample = SUB16(tfrc_rx_hist_rtt_prev_s(h)->tfrchrx_ccval,
-				       tfrc_rx_hist_rtt_last_s(h)->tfrchrx_ccval);
-			if (sample)
-				sample = 4 / sample *
-				         ktime_us_delta(tfrc_rx_hist_rtt_prev_s(h)->tfrchrx_tstamp,
-							tfrc_rx_hist_rtt_last_s(h)->tfrchrx_tstamp);
-			else    /*
-				 * FIXME: This condition is in principle not
-				 * possible but occurs when CCID is used for
-				 * two-way data traffic. I have tried to trace
-				 * it, but the cause does not seem to be here.
-				 */
-				DCCP_BUG("please report to dccp@vger.kernel.org"
-					 " => prev = %u, last = %u",
-					 tfrc_rx_hist_rtt_prev_s(h)->tfrchrx_ccval,
-					 tfrc_rx_hist_rtt_last_s(h)->tfrchrx_ccval);
-		} else if (delta_v < 1) {
-			h->rtt_sample_prev = 1;
-			goto keep_ref_for_next_time;
-		}
+	struct tfrc_rx_hist_entry *last = h->ring[0];
+	u32 sample, delta_v;
 
-	} else if (delta_v == 4) /* optimal match */
-		sample = ktime_to_us(net_timedelta(tfrc_rx_hist_rtt_last_s(h)->tfrchrx_tstamp));
-	else {			 /* suboptimal match */
-		h->rtt_sample_prev = 2;
-		goto keep_ref_for_next_time;
-	}
+	/*
+	 * When not to sample:
+	 * - on non-data packets
+	 *   (RFC 4342, 8.1: CCVal only fully defined for data packets);
+	 * - when no data packets have been received yet
+	 *   (FIXME: using sampled packet size as indicator here);
+	 * - as long as there are gaps in the sequence space (pending loss).
+	 */
+	if (!dccp_data_packet(skb) || h->packet_size == 0 ||
+	    tfrc_rx_hist_loss_pending(h))
+		return;
 
-	if (unlikely(sample > DCCP_SANE_RTT_MAX)) {
-		DCCP_WARN("RTT sample %u too large, using max\n", sample);
-		sample = DCCP_SANE_RTT_MAX;
+	h->rtt_sample_prev = 0;		/* reset previous candidate */
+
+	delta_v = SUB16(dccp_hdr(skb)->dccph_ccval, last->tfrchrx_ccval);
+	if (delta_v == 0) {		/* less than RTT/4 difference */
+		h->rtt_sample_prev = 1;
+		return;
 	}
+	sample = dccp_sane_rtt(ktime_to_us(net_timedelta(last->tfrchrx_tstamp)));
 
-	h->rtt_sample_prev = 0;	       /* use current entry as next reference */
-keep_ref_for_next_time:
+	if (delta_v <= 4)		/* between RTT/4 and RTT */
+		sample *= 4 / delta_v;
+	else if (!(sample < h->rtt_estimate && sample > h->rtt_estimate/2))
+		/*
+		* Optimisation: CCVal difference is greater than 1 RTT, yet the
+		* sample is less than the local RTT estimate; which means that
+		* the RTT estimate is too high.
+		* To avoid noise, it is not done if the sample is below RTT/2.
+		*/
+		return;
 
-	return sample;
+	/* Use a lower weight than usual to increase responsiveness */
+	h->rtt_estimate = tfrc_ewma(h->rtt_estimate, sample, 5);
 }
 EXPORT_SYMBOL_GPL(tfrc_rx_hist_sample_rtt);

net/dccp/ccids/lib/packet_history.h

@@ -40,12 +40,28 @@
 #include <linux/slab.h>
 #include "tfrc.h"
 
-struct tfrc_tx_hist_entry;
+/**
+ *  tfrc_tx_hist_entry  -  Simple singly-linked TX history list
+ *  @next:  next oldest entry (LIFO order)
+ *  @seqno: sequence number of this entry
+ *  @stamp: send time of packet with sequence number @seqno
+ */
+struct tfrc_tx_hist_entry {
+	struct tfrc_tx_hist_entry *next;
+	u64			  seqno;
+	ktime_t			  stamp;
+};
+
+static inline struct tfrc_tx_hist_entry *
+	tfrc_tx_hist_find_entry(struct tfrc_tx_hist_entry *head, u64 seqno)
+{
+	while (head != NULL && head->seqno != seqno)
+		head = head->next;
+	return head;
+}
 
 extern int  tfrc_tx_hist_add(struct tfrc_tx_hist_entry **headp, u64 seqno);
 extern void tfrc_tx_hist_purge(struct tfrc_tx_hist_entry **headp);
-extern u32  tfrc_tx_hist_rtt(struct tfrc_tx_hist_entry *head,
-			     const u64 seqno, const ktime_t now);
 
 /* Subtraction a-b modulo-16, respects circular wrap-around */
 #define SUB16(a, b) (((a) + 16 - (b)) & 0xF)
@@ -75,12 +91,22 @@ struct tfrc_rx_hist_entry {
  * @loss_count:		Number of entries in circular history
  * @loss_start:		Movable index (for loss detection)
  * @rtt_sample_prev:	Used during RTT sampling, points to candidate entry
+ * @rtt_estimate:	Receiver RTT estimate
+ * @packet_size:	Packet size in bytes (as per RFC 3448, 3.1)
+ * @bytes_recvd:	Number of bytes received since @bytes_start
+ * @bytes_start:	Start time for counting @bytes_recvd
  */
 struct tfrc_rx_hist {
 	struct tfrc_rx_hist_entry *ring[TFRC_NDUPACK + 1];
 	u8			  loss_count:2,
 				  loss_start:2;
+	/* Receiver RTT sampling */
 #define rtt_sample_prev		  loss_start
+	u32			  rtt_estimate;
+	/* Receiver sampling of application payload lengths */
+	u32			  packet_size,
+				  bytes_recvd;
+	ktime_t			  bytes_start;
 };
 
 /**
@@ -124,20 +150,50 @@ static inline bool tfrc_rx_hist_loss_pending(const struct tfrc_rx_hist *h)
 	return h->loss_count > 0;
 }
 
+/*
+ * Accessor functions to retrieve parameters sampled by the RX history
+ */
+static inline u32 tfrc_rx_hist_packet_size(const struct tfrc_rx_hist *h)
+{
+	if (h->packet_size == 0) {
+		DCCP_WARN("No sample for s, using fallback\n");
+		return TCP_MIN_RCVMSS;
+	}
+	return h->packet_size;
+
+}
+static inline u32 tfrc_rx_hist_rtt(const struct tfrc_rx_hist *h)
+{
+	if (h->rtt_estimate == 0) {
+		DCCP_WARN("No RTT estimate available, using fallback RTT\n");
+		return  DCCP_FALLBACK_RTT;
+	}
+	return h->rtt_estimate;
+}
+
+static inline void tfrc_rx_hist_restart_byte_counter(struct tfrc_rx_hist *h)
+{
+	h->bytes_recvd = 0;
+	h->bytes_start = ktime_get_real();
+}
+
+extern u32  tfrc_rx_hist_x_recv(struct tfrc_rx_hist *h, const u32 last_x_recv);
+
+
 extern void tfrc_rx_hist_add_packet(struct tfrc_rx_hist *h,
 				    const struct sk_buff *skb, const u64 ndp);
 
 extern int tfrc_rx_hist_duplicate(struct tfrc_rx_hist *h, struct sk_buff *skb);
 
 struct tfrc_loss_hist;
-extern int  tfrc_rx_handle_loss(struct tfrc_rx_hist *h,
-				struct tfrc_loss_hist *lh,
-				struct sk_buff *skb, const u64 ndp,
-				u32 (*first_li)(struct sock *sk),
-				struct sock *sk);
-extern u32 tfrc_rx_hist_sample_rtt(struct tfrc_rx_hist *h,
-				   const struct sk_buff *skb);
-extern int tfrc_rx_hist_alloc(struct tfrc_rx_hist *h);
+extern bool tfrc_rx_congestion_event(struct tfrc_rx_hist *h,
+				     struct tfrc_loss_hist *lh,
+				     struct sk_buff *skb, const u64 ndp,
+				     u32 (*first_li)(struct sock *sk),
+				     struct sock *sk);
+extern void tfrc_rx_hist_sample_rtt(struct tfrc_rx_hist *h,
+				    const struct sk_buff *skb);
+extern int  tfrc_rx_hist_init(struct tfrc_rx_hist *h, struct sock *sk);
 extern void tfrc_rx_hist_purge(struct tfrc_rx_hist *h);
 
 #endif /* _DCCP_PKT_HIST_ */

net/dccp/ccids/lib/tfrc.h

@@ -47,6 +47,21 @@ static inline u32 scaled_div32(u64 a, u64 b)
 	return result;
 }
 
+/**
+ * tfrc_scaled_sqrt  -  Compute scaled integer sqrt(x) for 0 < x < 2^22-1
+ * Uses scaling to improve accuracy of the integer approximation of sqrt(). The
+ * scaling factor of 2^10 limits the maximum @sample to 4e6; this is okay for
+ * clamped RTT samples (dccp_sample_rtt).
+ * Should best be used for expressions of type sqrt(x)/sqrt(y), since then the
+ * scaling factor is neutralised. For this purpose, it avoids returning zero.
+ */
+static inline u16 tfrc_scaled_sqrt(const u32 sample)
+{
+	const unsigned long non_zero_sample = sample ? : 1;
+
+	return int_sqrt(non_zero_sample << 10);
+}
+
 /**
  * tfrc_ewma  -  Exponentially weighted moving average
  * @weight: Weight to be used as damping factor, in units of 1/10
@@ -58,6 +73,7 @@ static inline u32 tfrc_ewma(const u32 avg, const u32 newval, const u8 weight)
 
 extern u32  tfrc_calc_x(u16 s, u32 R, u32 p);
 extern u32  tfrc_calc_x_reverse_lookup(u32 fvalue);
+extern u32  tfrc_invert_loss_event_rate(u32 loss_event_rate);
 
 extern int  tfrc_tx_packet_history_init(void);
 extern void tfrc_tx_packet_history_exit(void);

net/dccp/ccids/lib/tfrc_equation.c

@@ -632,8 +632,16 @@ u32 tfrc_calc_x(u16 s, u32 R, u32 p)
 
 	if (p <= TFRC_CALC_X_SPLIT) 		{     /* 0.0000 < p <= 0.05   */
 		if (p < TFRC_SMALLEST_P) {	      /* 0.0000 < p <  0.0001 */
-			DCCP_WARN("Value of p (%d) below resolution. "
-				  "Substituting %d\n", p, TFRC_SMALLEST_P);
+			/*
+			 * In the congestion-avoidance phase p decays towards 0
+			 * when there are no further losses, so this case is
+			 * natural. Truncating to p_min = 0.01% means that the
+			 * maximum achievable throughput is limited to about
+			 * X_calc_max = 122.4 * s/RTT (see RFC 3448, 3.1); e.g.
+			 * with s=1500 bytes, RTT=0.01 s: X_calc_max = 147 Mbps.
+			 */
+			tfrc_pr_debug("Value of p (%d) below resolution. "
+				      "Substituting %d\n", p, TFRC_SMALLEST_P);
 			index = 0;
 		} else 				      /* 0.0001 <= p <= 0.05  */
 			index =  p/TFRC_SMALLEST_P - 1;
@@ -658,7 +666,6 @@ u32 tfrc_calc_x(u16 s, u32 R, u32 p)
 	result = scaled_div(s, R);
 	return scaled_div32(result, f);
 }
-
 EXPORT_SYMBOL_GPL(tfrc_calc_x);
 
 /**
@@ -693,5 +700,19 @@ u32 tfrc_calc_x_reverse_lookup(u32 fvalue)
 	index = tfrc_binsearch(fvalue, 0);
 	return (index + 1) * 1000000 / TFRC_CALC_X_ARRSIZE;
 }
-
 EXPORT_SYMBOL_GPL(tfrc_calc_x_reverse_lookup);
+
+/**
+ * tfrc_invert_loss_event_rate  -  Compute p so that 10^6 corresponds to 100%
+ * When @loss_event_rate is large, there is a chance that p is truncated to 0.
+ * To avoid re-entering slow-start in that case, we set p = TFRC_SMALLEST_P > 0.
+ */
+u32 tfrc_invert_loss_event_rate(u32 loss_event_rate)
+{
+	if (loss_event_rate == UINT_MAX)		/* see RFC 4342, 8.5 */
+		return 0;
+	if (unlikely(loss_event_rate == 0))		/* map 1/0 into 100% */
+		return 1000000;
+	return max_t(u32, scaled_div(1, loss_event_rate), TFRC_SMALLEST_P);
+}
+EXPORT_SYMBOL_GPL(tfrc_invert_loss_event_rate);

net/dccp/dccp.h

@@ -42,9 +42,11 @@
 extern int dccp_debug;
 #define dccp_pr_debug(format, a...)	  DCCP_PR_DEBUG(dccp_debug, format, ##a)
 #define dccp_pr_debug_cat(format, a...)   DCCP_PRINTK(dccp_debug, format, ##a)
+#define dccp_debug(fmt, a...)		  dccp_pr_debug_cat(KERN_DEBUG fmt, ##a)
 #else
 #define dccp_pr_debug(format, a...)
 #define dccp_pr_debug_cat(format, a...)
+#define dccp_debug(format, a...)
 #endif
 
 extern struct inet_hashinfo dccp_hashinfo;
@@ -61,11 +63,14 @@ extern void dccp_time_wait(struct sock *sk, int state, int timeo);
  *    - DCCP-Reset    with ACK Subheader and 4 bytes of Reset Code fields
  *  Hence a safe upper bound for the maximum option length is 1020-28 = 992
  */
-#define MAX_DCCP_SPECIFIC_HEADER (255 * sizeof(int))
+#define MAX_DCCP_SPECIFIC_HEADER (255 * sizeof(uint32_t))
 #define DCCP_MAX_PACKET_HDR 28
 #define DCCP_MAX_OPT_LEN (MAX_DCCP_SPECIFIC_HEADER - DCCP_MAX_PACKET_HDR)
 #define MAX_DCCP_HEADER (MAX_DCCP_SPECIFIC_HEADER + MAX_HEADER)
 
+/* Upper bound for initial feature-negotiation overhead (padded to 32 bits) */
+#define DCCP_FEATNEG_OVERHEAD	 (32 * sizeof(uint32_t))
+
 #define DCCP_TIMEWAIT_LEN (60 * HZ) /* how long to wait to destroy TIME-WAIT
 				     * state, about 60 seconds */
 
@@ -81,10 +86,13 @@ extern void dccp_time_wait(struct sock *sk, int state, int timeo);
  */
 #define DCCP_RTO_MAX ((unsigned)(64 * HZ))
 
+/* DCCP base time resolution - 10 microseconds (RFC 4340, 13.1 ... 13.3) */
+#define DCCP_TIME_RESOLUTION	10
+
 /*
  * RTT sampling: sanity bounds and fallback RTT value from RFC 4340, section 3.4
  */
-#define DCCP_SANE_RTT_MIN	100
+#define DCCP_SANE_RTT_MIN	(10 * DCCP_TIME_RESOLUTION)
 #define DCCP_FALLBACK_RTT	(USEC_PER_SEC / 5)
 #define DCCP_SANE_RTT_MAX	(3 * USEC_PER_SEC)
 
@@ -95,12 +103,6 @@ extern void dccp_time_wait(struct sock *sk, int state, int timeo);
 extern int  sysctl_dccp_request_retries;
 extern int  sysctl_dccp_retries1;
 extern int  sysctl_dccp_retries2;
-extern int  sysctl_dccp_feat_sequence_window;
-extern int  sysctl_dccp_feat_rx_ccid;
-extern int  sysctl_dccp_feat_tx_ccid;
-extern int  sysctl_dccp_feat_ack_ratio;
-extern int  sysctl_dccp_feat_send_ack_vector;
-extern int  sysctl_dccp_feat_send_ndp_count;
 extern int  sysctl_dccp_tx_qlen;
 extern int  sysctl_dccp_sync_ratelimit;
 
@@ -235,8 +237,22 @@ extern void dccp_reqsk_send_ack(struct sock *sk, struct sk_buff *skb,
 extern void dccp_send_sync(struct sock *sk, const u64 seq,
 			   const enum dccp_pkt_type pkt_type);
 
-extern void dccp_write_xmit(struct sock *sk, int block);
+/*
+ * TX Packet Dequeueing Interface
+ */
+extern void		dccp_qpolicy_push(struct sock *sk, struct sk_buff *skb);
+extern bool		dccp_qpolicy_full(struct sock *sk);
+extern void		dccp_qpolicy_drop(struct sock *sk, struct sk_buff *skb);
+extern struct sk_buff	*dccp_qpolicy_top(struct sock *sk);
+extern struct sk_buff	*dccp_qpolicy_pop(struct sock *sk);
+extern bool		dccp_qpolicy_param_ok(struct sock *sk, __be32 param);
+
+/*
+ * TX Packet Output and TX Timers
+ */
+extern void dccp_write_xmit(struct sock *sk);
 extern void dccp_write_space(struct sock *sk);
+extern void dccp_flush_write_queue(struct sock *sk, long *time_budget);
 
 extern void dccp_init_xmit_timers(struct sock *sk);
 static inline void dccp_clear_xmit_timers(struct sock *sk)
@@ -252,7 +268,8 @@ extern const char *dccp_state_name(const int state);
 extern void dccp_set_state(struct sock *sk, const int state);
 extern void dccp_done(struct sock *sk);
 
-extern void dccp_reqsk_init(struct request_sock *req, struct sk_buff *skb);
+extern int  dccp_reqsk_init(struct request_sock *rq, struct dccp_sock const *dp,
+			    struct sk_buff const *skb);
 
 extern int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb);
 
@@ -317,7 +334,14 @@ extern struct sk_buff *dccp_ctl_make_reset(struct sock *sk,
 extern int	   dccp_send_reset(struct sock *sk, enum dccp_reset_codes code);
 extern void	   dccp_send_close(struct sock *sk, const int active);
 extern int	   dccp_invalid_packet(struct sk_buff *skb);
-extern u32	   dccp_sample_rtt(struct sock *sk, long delta);
+
+static inline u32  dccp_sane_rtt(long usec_sample)
+{
+	if (unlikely(usec_sample <= 0 || usec_sample > DCCP_SANE_RTT_MAX))
+		DCCP_WARN("RTT sample %ld out of bounds!\n", usec_sample);
+	return clamp_val(usec_sample, DCCP_SANE_RTT_MIN, DCCP_SANE_RTT_MAX);
+}
+extern u32 dccp_sample_rtt(struct sock *sk, long delta);
 
 static inline int dccp_bad_service_code(const struct sock *sk,
 					const __be32 service)
@@ -411,36 +435,62 @@ static inline void dccp_hdr_set_ack(struct dccp_hdr_ack_bits *dhack,
 static inline void dccp_update_gsr(struct sock *sk, u64 seq)
 {
 	struct dccp_sock *dp = dccp_sk(sk);
-	const struct dccp_minisock *dmsk = dccp_msk(sk);
 
 	dp->dccps_gsr = seq;
-	dccp_set_seqno(&dp->dccps_swl,
-		       dp->dccps_gsr + 1 - (dmsk->dccpms_sequence_window / 4));
-	dccp_set_seqno(&dp->dccps_swh,
-		       dp->dccps_gsr + (3 * dmsk->dccpms_sequence_window) / 4);
+	/* Sequence validity window depends on remote Sequence Window (7.5.1) */
+	dp->dccps_swl = SUB48(ADD48(dp->dccps_gsr, 1), dp->dccps_r_seq_win / 4);
+	/*
+	 * Adjust SWL so that it is not below ISR. In contrast to RFC 4340,
+	 * 7.5.1 we perform this check beyond the initial handshake: W/W' are
+	 * always > 32, so for the first W/W' packets in the lifetime of a
+	 * connection we always have to adjust SWL.
+	 * A second reason why we are doing this is that the window depends on
+	 * the feature-remote value of Sequence Window: nothing stops the peer
+	 * from updating this value while we are busy adjusting SWL for the
+	 * first W packets (we would have to count from scratch again then).
+	 * Therefore it is safer to always make sure that the Sequence Window
+	 * is not artificially extended by a peer who grows SWL downwards by
+	 * continually updating the feature-remote Sequence-Window.
+	 * If sequence numbers wrap it is bad luck. But that will take a while
+	 * (48 bit), and this measure prevents Sequence-number attacks.
+	 */
+	if (before48(dp->dccps_swl, dp->dccps_isr))
+		dp->dccps_swl = dp->dccps_isr;
+	dp->dccps_swh = ADD48(dp->dccps_gsr, (3 * dp->dccps_r_seq_win) / 4);
 }
 
 static inline void dccp_update_gss(struct sock *sk, u64 seq)
 {
 	struct dccp_sock *dp = dccp_sk(sk);
 
-	dp->dccps_awh = dp->dccps_gss = seq;
-	dccp_set_seqno(&dp->dccps_awl,
-		       (dp->dccps_gss -
-			dccp_msk(sk)->dccpms_sequence_window + 1));
+	dp->dccps_gss = seq;
+	/* Ack validity window depends on local Sequence Window value (7.5.1) */
+	dp->dccps_awl = SUB48(ADD48(dp->dccps_gss, 1), dp->dccps_l_seq_win);
+	/* Adjust AWL so that it is not below ISS - see comment above for SWL */
+	if (before48(dp->dccps_awl, dp->dccps_iss))
+		dp->dccps_awl = dp->dccps_iss;
+	dp->dccps_awh = dp->dccps_gss;
+}
+
+static inline int dccp_ackvec_pending(const struct sock *sk)
+{
+	return dccp_sk(sk)->dccps_hc_rx_ackvec != NULL &&
+	       !dccp_ackvec_is_empty(dccp_sk(sk)->dccps_hc_rx_ackvec);
 }
 
 static inline int dccp_ack_pending(const struct sock *sk)
 {
-	const struct dccp_sock *dp = dccp_sk(sk);
-	return dp->dccps_timestamp_echo != 0 ||
-#ifdef CONFIG_IP_DCCP_ACKVEC
-	       (dccp_msk(sk)->dccpms_send_ack_vector &&
-		dccp_ackvec_pending(dp->dccps_hc_rx_ackvec)) ||
-#endif
-	       inet_csk_ack_scheduled(sk);
+	return dccp_ackvec_pending(sk) || inet_csk_ack_scheduled(sk);
 }
 
+extern int  dccp_feat_signal_nn_change(struct sock *sk, u8 feat, u64 nn_val);
+extern int  dccp_feat_finalise_settings(struct dccp_sock *dp);
+extern int  dccp_feat_server_ccid_dependencies(struct dccp_request_sock *dreq);
+extern int  dccp_feat_insert_opts(struct dccp_sock*, struct dccp_request_sock*,
+				  struct sk_buff *skb);
+extern int  dccp_feat_activate_values(struct sock *sk, struct list_head *fn);
+extern void dccp_feat_list_purge(struct list_head *fn_list);
+
 extern int dccp_insert_options(struct sock *sk, struct sk_buff *skb);
 extern int dccp_insert_options_rsk(struct dccp_request_sock*, struct sk_buff*);
 extern int dccp_insert_option_elapsed_time(struct sock *sk,

net/dccp/diag.c

@@ -29,7 +29,7 @@ static void dccp_get_info(struct sock *sk, struct tcp_info *info)
 	info->tcpi_backoff	= icsk->icsk_backoff;
 	info->tcpi_pmtu		= icsk->icsk_pmtu_cookie;
 
-	if (dccp_msk(sk)->dccpms_send_ack_vector)
+	if (dp->dccps_hc_rx_ackvec != NULL)
 		info->tcpi_options |= TCPI_OPT_SACK;
 
 	ccid_hc_rx_get_info(dp->dccps_hc_rx_ccid, sk, info);

net/dccp/feat.h

@@ -3,38 +3,134 @@
 /*
  *  net/dccp/feat.h
  *
- *  An implementation of the DCCP protocol
+ *  Feature negotiation for the DCCP protocol (RFC 4340, section 6)
+ *  Copyright (c) 2008 Gerrit Renker <gerrit@erg.abdn.ac.uk>
  *  Copyright (c) 2005 Andrea Bittau <a.bittau@cs.ucl.ac.uk>
  *
- *	This program is free software; you can redistribute it and/or modify it
- *	under the terms of the GNU General Public License version 2 as
- *	published by the Free Software Foundation.
+ *  This program is free software; you can redistribute it and/or modify it
+ *  under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
  */
-
 #include <linux/types.h>
 #include "dccp.h"
 
-#ifdef CONFIG_IP_DCCP_DEBUG
-extern const char *dccp_feat_typename(const u8 type);
-extern const char *dccp_feat_name(const u8 feat);
+/*
+ * Known limit values
+ */
+/* Ack Ratio takes 2-byte integer values (11.3) */
+#define DCCPF_ACK_RATIO_MAX	0xFFFF
+/* Wmin=32 and Wmax=2^46-1 from 7.5.2 */
+#define DCCPF_SEQ_WMIN		32
+#define DCCPF_SEQ_WMAX		0x3FFFFFFFFFFFull
+/* Maximum number of SP values that fit in a single (Confirm) option */
+#define DCCP_FEAT_MAX_SP_VALS	(DCCP_SINGLE_OPT_MAXLEN - 2)
+
+enum dccp_feat_type {
+	FEAT_AT_RX   = 1,	/* located at RX side of half-connection  */
+	FEAT_AT_TX   = 2,	/* located at TX side of half-connection  */
+	FEAT_SP      = 4,	/* server-priority reconciliation (6.3.1) */
+	FEAT_NN	     = 8,	/* non-negotiable reconciliation (6.3.2)  */
+	FEAT_UNKNOWN = 0xFF	/* not understood or invalid feature	  */
+};
+
+enum dccp_feat_state {
+	FEAT_DEFAULT = 0,	/* using default values from 6.4 */
+	FEAT_INITIALISING,	/* feature is being initialised  */
+	FEAT_CHANGING,		/* Change sent but not confirmed yet */
+	FEAT_UNSTABLE,		/* local modification in state CHANGING */
+	FEAT_STABLE		/* both ends (think they) agree */
+};
 
-static inline void dccp_feat_debug(const u8 type, const u8 feat, const u8 val)
+/**
+ * dccp_feat_val  -  Container for SP or NN feature values
+ * @nn:     single NN value
+ * @sp.vec: single SP value plus optional preference list
+ * @sp.len: length of @sp.vec in bytes
+ */
+typedef union {
+	u64 nn;
+	struct {
+		u8	*vec;
+		u8	len;
+	}   sp;
+} dccp_feat_val;
+
+/**
+ * struct feat_entry  -  Data structure to perform feature negotiation
+ * @feat_num: one of %dccp_feature_numbers
+ * @val: feature's current value (SP features may have preference list)
+ * @state: feature's current state
+ * @needs_mandatory: whether Mandatory options should be sent
+ * @needs_confirm: whether to send a Confirm instead of a Change
+ * @empty_confirm: whether to send an empty Confirm (depends on @needs_confirm)
+ * @is_local: feature location (1) or feature-remote (0)
+ * @node: list pointers, entries arranged in FIFO order
+ */
+struct dccp_feat_entry {
+	u8                      feat_num;
+	dccp_feat_val           val;
+	enum dccp_feat_state    state:8;
+	bool			needs_mandatory:1,
+				needs_confirm:1,
+				empty_confirm:1,
+				is_local:1;
+
+	struct list_head	node;
+};
+
+static inline u8 dccp_feat_genopt(struct dccp_feat_entry *entry)
 {
-	dccp_pr_debug("%s(%s (%d), %d)\n", dccp_feat_typename(type),
-					   dccp_feat_name(feat), feat, val);
+	if (entry->needs_confirm)
+		return entry->is_local ? DCCPO_CONFIRM_L : DCCPO_CONFIRM_R;
+	return entry->is_local ? DCCPO_CHANGE_L : DCCPO_CHANGE_R;
 }
-#else
-#define dccp_feat_debug(type, feat, val)
-#endif /* CONFIG_IP_DCCP_DEBUG */
-
-extern int  dccp_feat_change(struct dccp_minisock *dmsk, u8 type, u8 feature,
-			     u8 *val, u8 len, gfp_t gfp);
-extern int  dccp_feat_change_recv(struct sock *sk, u8 type, u8 feature,
-				  u8 *val, u8 len);
-extern int  dccp_feat_confirm_recv(struct sock *sk, u8 type, u8 feature,
-				   u8 *val, u8 len);
-extern void dccp_feat_clean(struct dccp_minisock *dmsk);
-extern int  dccp_feat_clone(struct sock *oldsk, struct sock *newsk);
-extern int  dccp_feat_init(struct dccp_minisock *dmsk);
 
+/**
+ * struct ccid_dependency  -  Track changes resulting from choosing a CCID
+ * @dependent_feat: one of %dccp_feature_numbers
+ * @is_local: local (1) or remote (0) @dependent_feat
+ * @is_mandatory: whether presence of @dependent_feat is mission-critical or not
+ * @val: corresponding default value for @dependent_feat (u8 is sufficient here)
+ */
+struct ccid_dependency {
+	u8	dependent_feat;
+	bool	is_local:1,
+		is_mandatory:1;
+	u8	val;
+};
+
+/*
+ * Sysctls to seed defaults for feature negotiation
+ */
+extern unsigned long sysctl_dccp_sequence_window;
+extern int	     sysctl_dccp_rx_ccid;
+extern int	     sysctl_dccp_tx_ccid;
+
+extern int  dccp_feat_init(struct sock *sk);
+extern void dccp_feat_initialise_sysctls(void);
+extern int  dccp_feat_register_sp(struct sock *sk, u8 feat, u8 is_local,
+				  u8 const *list, u8 len);
+extern int  dccp_feat_register_nn(struct sock *sk, u8 feat, u64 val);
+extern int  dccp_feat_parse_options(struct sock *, struct dccp_request_sock *,
+				    u8 mand, u8 opt, u8 feat, u8 *val, u8 len);
+extern int  dccp_feat_clone_list(struct list_head const *, struct list_head *);
+
+/*
+ * Encoding variable-length options and their maximum length.
+ *
+ * This affects NN options (SP options are all u8) and other variable-length
+ * options (see table 3 in RFC 4340). The limit is currently given the Sequence
+ * Window NN value (sec. 7.5.2) and the NDP count (sec. 7.7) option, all other
+ * options consume less than 6 bytes (timestamps are 4 bytes).
+ * When updating this constant (e.g. due to new internet drafts / RFCs), make
+ * sure that you also update all code which refers to it.
+ */
+#define DCCP_OPTVAL_MAXLEN	6
+
+extern void dccp_encode_value_var(const u64 value, u8 *to, const u8 len);
+extern u64  dccp_decode_value_var(const u8 *bf, const u8 len);
+
+extern int  dccp_insert_option_mandatory(struct sk_buff *skb);
+extern int  dccp_insert_fn_opt(struct sk_buff *skb, u8 type, u8 feat,
+			       u8 *val, u8 len, bool repeat_first);
 #endif /* _DCCP_FEAT_H */

net/dccp/ipv4.c

@@ -545,6 +545,7 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
 
 static void dccp_v4_reqsk_destructor(struct request_sock *req)
 {
+	dccp_feat_list_purge(&dccp_rsk(req)->dreq_featneg);
 	kfree(inet_rsk(req)->opt);
 }
 
@@ -595,7 +596,8 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 	if (req == NULL)
 		goto drop;
 
-	dccp_reqsk_init(req, skb);
+	if (dccp_reqsk_init(req, dccp_sk(sk), skb))
+		goto drop_and_free;
 
 	dreq = dccp_rsk(req);
 	if (dccp_parse_options(sk, dreq, skb))

net/dccp/ipv6.c

@@ -302,6 +302,7 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req)
 
 static void dccp_v6_reqsk_destructor(struct request_sock *req)
 {
+	dccp_feat_list_purge(&dccp_rsk(req)->dreq_featneg);
 	if (inet6_rsk(req)->pktopts != NULL)
 		kfree_skb(inet6_rsk(req)->pktopts);
 }
@@ -424,7 +425,8 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 	if (req == NULL)
 		goto drop;
 
-	dccp_reqsk_init(req, skb);
+	if (dccp_reqsk_init(req, dccp_sk(sk), skb))
+		goto drop_and_free;
 
 	dreq = dccp_rsk(req);
 	if (dccp_parse_options(sk, dreq, skb))