From 2a91aa3967398fb94eccc8da67c82bce9f67afdf Mon Sep 17 00:00:00 2001 From: Andrea Bittau Date: Mon, 20 Mar 2006 17:41:47 -0800 Subject: [DCCP] CCID2: Initial CCID2 (TCP-Like) implementation Original work by Andrea Bittau, Arnaldo Melo cleaned up and fixed several issues on the merge process. For now CCID2 was turned the default for all SOCK_DCCP connections, but this will be remedied soon with the merge of the feature negotiation code. Signed-off-by: Andrea Bittau Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/dccp.h | 8 +- net/dccp/Kconfig | 4 + net/dccp/ccids/Kconfig | 39 ++- net/dccp/ccids/Makefile | 4 + net/dccp/ccids/ccid2.c | 838 ++++++++++++++++++++++++++++++++++++++++++++++++ net/dccp/ccids/ccid2.h | 69 ++++ net/dccp/ipv4.c | 1 + 7 files changed, 957 insertions(+), 6 deletions(-) create mode 100644 net/dccp/ccids/ccid2.c create mode 100644 net/dccp/ccids/ccid2.h diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 088529f54965..268b4579d7e5 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -314,9 +314,9 @@ static inline unsigned int dccp_hdr_len(const struct sk_buff *skb) /* initial values for each feature */ #define DCCPF_INITIAL_SEQUENCE_WINDOW 100 -/* FIXME: for now we're using CCID 3 (TFRC) */ -#define DCCPF_INITIAL_CCID 3 -#define DCCPF_INITIAL_SEND_ACK_VECTOR 0 +/* FIXME: for now we're using CCID 2 (TCP-Like) */ +#define DCCPF_INITIAL_CCID 2 +#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 @@ -430,6 +430,8 @@ struct dccp_sock { struct timeval dccps_timestamp_time; __u32 dccps_timestamp_echo; __u32 dccps_packet_size; + __u16 dccps_l_ack_ratio; + __u16 dccps_r_ack_ratio; unsigned long dccps_ndp_count; __u32 dccps_mss_cache; struct dccp_options dccps_options; diff --git a/net/dccp/Kconfig b/net/dccp/Kconfig index 187ac182e24b..24a6981e209a 100644 --- a/net/dccp/Kconfig +++ b/net/dccp/Kconfig @@ -24,6 +24,10 @@ config INET_DCCP_DIAG def_tristate y if (IP_DCCP = y && INET_DIAG = y) def_tristate m +config IP_DCCP_ACKVEC + depends on IP_DCCP + def_bool N + source "net/dccp/ccids/Kconfig" menu "DCCP Kernel Hacking" diff --git a/net/dccp/ccids/Kconfig b/net/dccp/ccids/Kconfig index 7684d83946a4..422af197171d 100644 --- a/net/dccp/ccids/Kconfig +++ b/net/dccp/ccids/Kconfig @@ -1,6 +1,34 @@ menu "DCCP CCIDs Configuration (EXPERIMENTAL)" depends on IP_DCCP && EXPERIMENTAL +config IP_DCCP_CCID2 + tristate "CCID2 (TCP) (EXPERIMENTAL)" + depends on IP_DCCP + select IP_DCCP_ACKVEC + ---help--- + CCID 2, TCP-like Congestion Control, denotes Additive Increase, + Multiplicative Decrease (AIMD) congestion control with behavior + modelled directly on TCP, including congestion window, slow start, + timeouts, and so forth [RFC 2581]. CCID 2 achieves maximum + bandwidth over the long term, consistent with the use of end-to-end + congestion control, but halves its congestion window in response to + each congestion event. This leads to the abrupt rate changes + typical of TCP. Applications should use CCID 2 if they prefer + maximum bandwidth utilization to steadiness of rate. This is often + the case for applications that are not playing their data directly + to the user. For example, a hypothetical application that + transferred files over DCCP, using application-level retransmissions + for lost packets, would prefer CCID 2 to CCID 3. On-line games may + also prefer CCID 2. + + CCID 2 is further described in: + http://www.icir.org/kohler/dccp/draft-ietf-dccp-ccid2-10.txt + + This text was extracted from: + http://www.icir.org/kohler/dccp/draft-ietf-dccp-spec-13.txt + + If in doubt, say M. + config IP_DCCP_CCID3 tristate "CCID3 (TFRC) (EXPERIMENTAL)" depends on IP_DCCP @@ -15,10 +43,15 @@ config IP_DCCP_CCID3 suitable than CCID 2 for applications such streaming media where a relatively smooth sending rate is of importance. - CCID 3 is further described in [CCID 3 PROFILE]. The TFRC - congestion control algorithms were initially described in RFC 3448. + CCID 3 is further described in: + + http://www.icir.org/kohler/dccp/draft-ietf-dccp-ccid3-11.txt. + + The TFRC congestion control algorithms were initially described in + RFC 3448. - This text was extracted from draft-ietf-dccp-spec-11.txt. + This text was extracted from: + http://www.icir.org/kohler/dccp/draft-ietf-dccp-spec-13.txt If in doubt, say M. diff --git a/net/dccp/ccids/Makefile b/net/dccp/ccids/Makefile index 956f79f50743..438f20bccff7 100644 --- a/net/dccp/ccids/Makefile +++ b/net/dccp/ccids/Makefile @@ -2,4 +2,8 @@ obj-$(CONFIG_IP_DCCP_CCID3) += dccp_ccid3.o dccp_ccid3-y := ccid3.o +obj-$(CONFIG_IP_DCCP_CCID2) += dccp_ccid2.o + +dccp_ccid2-y := ccid2.o + obj-y += lib/ diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c new file mode 100644 index 000000000000..4a7b87512560 --- /dev/null +++ b/net/dccp/ccids/ccid2.c @@ -0,0 +1,838 @@ +/* + * net/dccp/ccids/ccid2.c + * + * Copyright (c) 2005, 2006 Andrea Bittau + * + * Changes to meet Linux coding standards, and DCCP infrastructure fixes. + * + * Copyright (c) 2006 Arnaldo Carvalho de Melo + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +/* + * This implementation should follow: draft-ietf-dccp-ccid2-10.txt + * + * BUGS: + * - sequence number wrapping + * - jiffies wrapping + */ + +#include +#include "../ccid.h" +#include "../dccp.h" +#include "ccid2.h" + +static int ccid2_debug; + +#if 0 +#define CCID2_DEBUG +#endif + +#ifdef CCID2_DEBUG +#define ccid2_pr_debug(format, a...) \ + do { if (ccid2_debug) \ + printk(KERN_DEBUG "%s: " format, __FUNCTION__, ##a); \ + } while (0) +#else +#define ccid2_pr_debug(format, a...) +#endif + +static const int ccid2_seq_len = 128; + +static inline struct ccid2_hc_tx_sock *ccid2_hc_tx_sk(const struct sock *sk) +{ + return dccp_sk(sk)->dccps_hc_tx_ccid_private; +} + +static inline struct ccid2_hc_rx_sock *ccid2_hc_rx_sk(const struct sock *sk) +{ + return dccp_sk(sk)->dccps_hc_rx_ccid_private; +} + +#ifdef CCID2_DEBUG +static void ccid2_hc_tx_check_sanity(const struct ccid2_hc_tx_sock *hctx) +{ + int len = 0; + struct ccid2_seq *seqp; + int pipe = 0; + + seqp = hctx->ccid2hctx_seqh; + + /* there is data in the chain */ + if (seqp != hctx->ccid2hctx_seqt) { + seqp = seqp->ccid2s_prev; + len++; + if (!seqp->ccid2s_acked) + pipe++; + + while (seqp != hctx->ccid2hctx_seqt) { + struct ccid2_seq *prev; + + prev = seqp->ccid2s_prev; + len++; + if (!prev->ccid2s_acked) + pipe++; + + /* packets are sent sequentially */ + BUG_ON(seqp->ccid2s_seq <= prev->ccid2s_seq); + BUG_ON(seqp->ccid2s_sent < prev->ccid2s_sent); + BUG_ON(len > ccid2_seq_len); + + seqp = prev; + } + } + + BUG_ON(pipe != hctx->ccid2hctx_pipe); + ccid2_pr_debug("len of chain=%d\n", len); + + do { + seqp = seqp->ccid2s_prev; + len++; + BUG_ON(len > ccid2_seq_len); + } while(seqp != hctx->ccid2hctx_seqh); + + BUG_ON(len != ccid2_seq_len); + ccid2_pr_debug("total len=%d\n", len); +} +#else +#define ccid2_hc_tx_check_sanity(hctx) do {} while (0) +#endif + +static int ccid2_hc_tx_send_packet(struct sock *sk, + struct sk_buff *skb, int len) +{ + struct ccid2_hc_tx_sock *hctx; + + switch (DCCP_SKB_CB(skb)->dccpd_type) { + case 0: /* XXX data packets from userland come through like this */ + case DCCP_PKT_DATA: + case DCCP_PKT_DATAACK: + break; + /* No congestion control on other packets */ + default: + return 0; + } + + hctx = ccid2_hc_tx_sk(sk); + + ccid2_pr_debug("pipe=%d cwnd=%d\n", hctx->ccid2hctx_pipe, + hctx->ccid2hctx_cwnd); + + if (hctx->ccid2hctx_pipe < hctx->ccid2hctx_cwnd) { + /* OK we can send... make sure previous packet was sent off */ + if (!hctx->ccid2hctx_sendwait) { + hctx->ccid2hctx_sendwait = 1; + return 0; + } + } + + return 100; /* XXX */ +} + +static void ccid2_change_l_ack_ratio(struct sock *sk, int val) +{ + struct dccp_sock *dp = dccp_sk(sk); + /* + * XXX I don't really agree with val != 2. If cwnd is 1, ack ratio + * should be 1... it shouldn't be allowed to become 2. + * -sorbo. + */ + if (val != 2) { + struct ccid2_hc_tx_sock *hctx = ccid2_hc_tx_sk(sk); + int max = hctx->ccid2hctx_cwnd / 2; + + /* round up */ + if (hctx->ccid2hctx_cwnd & 1) + max++; + + if (val > max) + val = max; + } + + ccid2_pr_debug("changing local ack ratio to %d\n", val); + WARN_ON(val <= 0); + dp->dccps_l_ack_ratio = val; +} + +static void ccid2_change_cwnd(struct sock *sk, int val) +{ + struct ccid2_hc_tx_sock *hctx = ccid2_hc_tx_sk(sk); + + if (val == 0) + val = 1; + + /* XXX do we need to change ack ratio? */ + ccid2_pr_debug("change cwnd to %d\n", val); + + BUG_ON(val < 1); + hctx->ccid2hctx_cwnd = val; +} + +static void ccid2_start_rto_timer(struct sock *sk); + +static void ccid2_hc_tx_rto_expire(unsigned long data) +{ + struct sock *sk = (struct sock *)data; + struct ccid2_hc_tx_sock *hctx = ccid2_hc_tx_sk(sk); + long s; + + /* XXX I don't think i'm locking correctly + * -sorbo. + */ + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { + sk_reset_timer(sk, &hctx->ccid2hctx_rtotimer, + jiffies + HZ / 5); + goto out; + } + + ccid2_pr_debug("RTO_EXPIRE\n"); + + ccid2_hc_tx_check_sanity(hctx); + + /* back-off timer */ + hctx->ccid2hctx_rto <<= 1; + + s = hctx->ccid2hctx_rto / HZ; + if (s > 60) + hctx->ccid2hctx_rto = 60 * HZ; + + ccid2_start_rto_timer(sk); + + /* adjust pipe, cwnd etc */ + hctx->ccid2hctx_pipe = 0; + hctx->ccid2hctx_ssthresh = hctx->ccid2hctx_cwnd >> 1; + if (hctx->ccid2hctx_ssthresh < 2) + hctx->ccid2hctx_ssthresh = 2; + ccid2_change_cwnd(sk, 1); + + /* clear state about stuff we sent */ + hctx->ccid2hctx_seqt = hctx->ccid2hctx_seqh; + hctx->ccid2hctx_ssacks = 0; + hctx->ccid2hctx_acks = 0; + hctx->ccid2hctx_sent = 0; + + /* clear ack ratio state. */ + hctx->ccid2hctx_arsent = 0; + hctx->ccid2hctx_ackloss = 0; + hctx->ccid2hctx_rpseq = 0; + hctx->ccid2hctx_rpdupack = -1; + ccid2_change_l_ack_ratio(sk, 1); + ccid2_hc_tx_check_sanity(hctx); +out: + bh_unlock_sock(sk); +/* sock_put(sk); */ +} + +static void ccid2_start_rto_timer(struct sock *sk) +{ + struct ccid2_hc_tx_sock *hctx = ccid2_hc_tx_sk(sk); + + ccid2_pr_debug("setting RTO timeout=%ld\n", hctx->ccid2hctx_rto); + + BUG_ON(timer_pending(&hctx->ccid2hctx_rtotimer)); + sk_reset_timer(sk, &hctx->ccid2hctx_rtotimer, + jiffies + hctx->ccid2hctx_rto); +} + +static void ccid2_hc_tx_packet_sent(struct sock *sk, int more, int len) +{ + struct dccp_sock *dp = dccp_sk(sk); + struct ccid2_hc_tx_sock *hctx = ccid2_hc_tx_sk(sk); + u64 seq; + + ccid2_hc_tx_check_sanity(hctx); + + BUG_ON(!hctx->ccid2hctx_sendwait); + hctx->ccid2hctx_sendwait = 0; + hctx->ccid2hctx_pipe++; + BUG_ON(hctx->ccid2hctx_pipe < 0); + + /* There is an issue. What if another packet is sent between + * packet_send() and packet_sent(). Then the sequence number would be + * wrong. + * -sorbo. + */ + seq = dp->dccps_gss; + + hctx->ccid2hctx_seqh->ccid2s_seq = seq; + hctx->ccid2hctx_seqh->ccid2s_acked = 0; + hctx->ccid2hctx_seqh->ccid2s_sent = jiffies; + hctx->ccid2hctx_seqh = hctx->ccid2hctx_seqh->ccid2s_next; + + ccid2_pr_debug("cwnd=%d pipe=%d\n", hctx->ccid2hctx_cwnd, + hctx->ccid2hctx_pipe); + + if (hctx->ccid2hctx_seqh == hctx->ccid2hctx_seqt) { + /* XXX allocate more space */ + WARN_ON(1); + } + + hctx->ccid2hctx_sent++; + + /* Ack Ratio. Need to maintain a concept of how many windows we sent */ + hctx->ccid2hctx_arsent++; + /* We had an ack loss in this window... */ + if (hctx->ccid2hctx_ackloss) { + if (hctx->ccid2hctx_arsent >= hctx->ccid2hctx_cwnd) { + hctx->ccid2hctx_arsent = 0; + hctx->ccid2hctx_ackloss = 0; + } + } + /* No acks lost up to now... */ + else { + /* decrease ack ratio if enough packets were sent */ + if (dp->dccps_l_ack_ratio > 1) { + /* XXX don't calculate denominator each time */ + int denom; + + denom = dp->dccps_l_ack_ratio * dp->dccps_l_ack_ratio - + dp->dccps_l_ack_ratio; + denom = hctx->ccid2hctx_cwnd * hctx->ccid2hctx_cwnd / denom; + + if (hctx->ccid2hctx_arsent >= denom) { + ccid2_change_l_ack_ratio(sk, dp->dccps_l_ack_ratio - 1); + hctx->ccid2hctx_arsent = 0; + } + } + /* we can't increase ack ratio further [1] */ + else { + hctx->ccid2hctx_arsent = 0; /* or maybe set it to cwnd*/ + } + } + + /* setup RTO timer */ + if (!timer_pending(&hctx->ccid2hctx_rtotimer)) { + ccid2_start_rto_timer(sk); + } +#ifdef CCID2_DEBUG + ccid2_pr_debug("pipe=%d\n", hctx->ccid2hctx_pipe); + ccid2_pr_debug("Sent: seq=%llu\n", seq); + do { + struct ccid2_seq *seqp = hctx->ccid2hctx_seqt; + + while (seqp != hctx->ccid2hctx_seqh) { + ccid2_pr_debug("out seq=%llu acked=%d time=%lu\n", + seqp->ccid2s_seq, seqp->ccid2s_acked, + seqp->ccid2s_sent); + seqp = seqp->ccid2s_next; + } + } while(0); + ccid2_pr_debug("=========\n"); + ccid2_hc_tx_check_sanity(hctx); +#endif +} + +/* XXX Lame code duplication! + * returns -1 if none was found. + * else returns the next offset to use in the function call. + */ +static int ccid2_ackvector(struct sock *sk, struct sk_buff *skb, int offset, + unsigned char **vec, unsigned char *veclen) +{ + const struct dccp_hdr *dh = dccp_hdr(skb); + unsigned char *options = (unsigned char *)dh + dccp_hdr_len(skb); + unsigned char *opt_ptr; + const unsigned char *opt_end = (unsigned char *)dh + + (dh->dccph_doff * 4); + unsigned char opt, len; + unsigned char *value; + + BUG_ON(offset < 0); + options += offset; + opt_ptr = options; + if (opt_ptr >= opt_end) + return -1; + + while (opt_ptr != opt_end) { + opt = *opt_ptr++; + len = 0; + value = NULL; + + /* Check if this isn't a single byte option */ + if (opt > DCCPO_MAX_RESERVED) { + if (opt_ptr == opt_end) + goto out_invalid_option; + + len = *opt_ptr++; + if (len < 3) + goto out_invalid_option; + /* + * Remove the type and len fields, leaving + * just the value size + */ + len -= 2; + value = opt_ptr; + opt_ptr += len; + + if (opt_ptr > opt_end) + goto out_invalid_option; + } + + switch (opt) { + case DCCPO_ACK_VECTOR_0: + case DCCPO_ACK_VECTOR_1: + *vec = value; + *veclen = len; + return offset + (opt_ptr - options); + break; + } + } + + return -1; + +out_invalid_option: + BUG_ON(1); /* should never happen... options were previously parsed ! */ + return -1; +} + +static void ccid2_hc_tx_kill_rto_timer(struct ccid2_hc_tx_sock *hctx) +{ + if (del_timer(&hctx->ccid2hctx_rtotimer)) + ccid2_pr_debug("deleted RTO timer\n"); +} + +static inline void ccid2_new_ack(struct sock *sk, + struct ccid2_seq *seqp, + unsigned int *maxincr) +{ + struct ccid2_hc_tx_sock *hctx = ccid2_hc_tx_sk(sk); + + /* slow start */ + if (hctx->ccid2hctx_cwnd < hctx->ccid2hctx_ssthresh) { + hctx->ccid2hctx_acks = 0; + + /* We can increase cwnd at most maxincr [ack_ratio/2] */ + if (*maxincr) { + /* increase every 2 acks */ + hctx->ccid2hctx_ssacks++; + if (hctx->ccid2hctx_ssacks == 2) { + ccid2_change_cwnd(sk, hctx->ccid2hctx_cwnd + 1); + hctx->ccid2hctx_ssacks = 0; + *maxincr = *maxincr - 1; + } + } + /* increased cwnd enough for this single ack */ + else { + hctx->ccid2hctx_ssacks = 0; + } + } + else { + hctx->ccid2hctx_ssacks = 0; + hctx->ccid2hctx_acks++; + + if (hctx->ccid2hctx_acks >= hctx->ccid2hctx_cwnd) { + ccid2_change_cwnd(sk, hctx->ccid2hctx_cwnd + 1); + hctx->ccid2hctx_acks = 0; + } + } + + /* update RTO */ + if (hctx->ccid2hctx_srtt == -1 || + (jiffies - hctx->ccid2hctx_lastrtt) >= hctx->ccid2hctx_srtt) { + unsigned long r = jiffies - seqp->ccid2s_sent; + int s; + + /* first measurement */ + if (hctx->ccid2hctx_srtt == -1) { + ccid2_pr_debug("R: %lu Time=%lu seq=%llu\n", + r, jiffies, seqp->ccid2s_seq); + hctx->ccid2hctx_srtt = r; + hctx->ccid2hctx_rttvar = r >> 1; + } + else { + /* RTTVAR */ + long tmp = hctx->ccid2hctx_srtt - r; + if (tmp < 0) + tmp *= -1; + + tmp >>= 2; + hctx->ccid2hctx_rttvar *= 3; + hctx->ccid2hctx_rttvar >>= 2; + hctx->ccid2hctx_rttvar += tmp; + + /* SRTT */ + hctx->ccid2hctx_srtt *= 7; + hctx->ccid2hctx_srtt >>= 3; + tmp = r >> 3; + hctx->ccid2hctx_srtt += tmp; + } + s = hctx->ccid2hctx_rttvar << 2; + /* clock granularity is 1 when based on jiffies */ + if (!s) + s = 1; + hctx->ccid2hctx_rto = hctx->ccid2hctx_srtt + s; + + /* must be at least a second */ + s = hctx->ccid2hctx_rto / HZ; + /* DCCP doesn't require this [but I like it cuz my code sux] */ +#if 1 + if (s < 1) + hctx->ccid2hctx_rto = HZ; +#endif + /* max 60 seconds */ + if (s > 60) + hctx->ccid2hctx_rto = HZ * 60; + + hctx->ccid2hctx_lastrtt = jiffies; + + ccid2_pr_debug("srtt: %ld rttvar: %ld rto: %ld (HZ=%d) R=%lu\n", + hctx->ccid2hctx_srtt, hctx->ccid2hctx_rttvar, + hctx->ccid2hctx_rto, HZ, r); + hctx->ccid2hctx_sent = 0; + } + + /* we got a new ack, so re-start RTO timer */ + ccid2_hc_tx_kill_rto_timer(hctx); + ccid2_start_rto_timer(sk); +} + +static void ccid2_hc_tx_dec_pipe(struct ccid2_hc_tx_sock *hctx) +{ + hctx->ccid2hctx_pipe--; + BUG_ON(hctx->ccid2hctx_pipe < 0); + + if (hctx->ccid2hctx_pipe == 0) + ccid2_hc_tx_kill_rto_timer(hctx); +} + +static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) +{ + struct dccp_sock *dp = dccp_sk(sk); + struct ccid2_hc_tx_sock *hctx = ccid2_hc_tx_sk(sk); + u64 ackno, seqno; + struct ccid2_seq *seqp; + unsigned char *vector; + unsigned char veclen; + int offset = 0; + int done = 0; + int loss = 0; + unsigned int maxincr = 0; + + ccid2_hc_tx_check_sanity(hctx); + /* check reverse path congestion */ + seqno = DCCP_SKB_CB(skb)->dccpd_seq; + + /* XXX this whole "algorithm" is broken. Need to fix it to keep track + * of the seqnos of the dupacks so that rpseq and rpdupack are correct + * -sorbo. + */ + /* need to bootstrap */ + if (hctx->ccid2hctx_rpdupack == -1) { + hctx->ccid2hctx_rpdupack = 0; + hctx->ccid2hctx_rpseq = seqno; + } + else { + /* check if packet is consecutive */ + if ((hctx->ccid2hctx_rpseq + 1) == seqno) { + hctx->ccid2hctx_rpseq++; + } + /* it's a later packet */ + else if (after48(seqno, hctx->ccid2hctx_rpseq)) { + hctx->ccid2hctx_rpdupack++; + + /* check if we got enough dupacks */ + if (hctx->ccid2hctx_rpdupack >= + hctx->ccid2hctx_numdupack) { + + hctx->ccid2hctx_rpdupack = -1; /* XXX lame */ + hctx->ccid2hctx_rpseq = 0; + + ccid2_change_l_ack_ratio(sk, dp->dccps_l_ack_ratio << 1); + } + } + } + + /* check forward path congestion */ + /* still didn't send out new data packets */ + if (hctx->ccid2hctx_seqh == hctx->ccid2hctx_seqt) + return; + + switch (DCCP_SKB_CB(skb)->dccpd_type) { + case DCCP_PKT_ACK: + case DCCP_PKT_DATAACK: + break; + + default: + return; + } + + ackno = DCCP_SKB_CB(skb)->dccpd_ack_seq; + seqp = hctx->ccid2hctx_seqh->ccid2s_prev; + + /* If in slow-start, cwnd can increase at most Ack Ratio / 2 packets for + * this single ack. I round up. + * -sorbo. + */ + maxincr = dp->dccps_l_ack_ratio >> 1; + maxincr++; + + /* go through all ack vectors */ + while ((offset = ccid2_ackvector(sk, skb, offset, + &vector, &veclen)) != -1) { + /* go through this ack vector */ + while (veclen--) { + const u8 rl = *vector & DCCP_ACKVEC_LEN_MASK; + u64 ackno_end_rl; + + dccp_set_seqno(&ackno_end_rl, ackno - rl); + ccid2_pr_debug("ackvec start:%llu end:%llu\n", ackno, + ackno_end_rl); + /* if the seqno we are analyzing is larger than the + * current ackno, then move towards the tail of our + * seqnos. + */ + while (after48(seqp->ccid2s_seq, ackno)) { + if (seqp == hctx->ccid2hctx_seqt) { + done = 1; + break; + } + seqp = seqp->ccid2s_prev; + } + if (done) + break; + + /* check all seqnos in the range of the vector + * run length + */ + while (between48(seqp->ccid2s_seq,ackno_end_rl,ackno)) { + const u8 state = (*vector & + DCCP_ACKVEC_STATE_MASK) >> 6; + + /* new packet received or marked */ + if (state != DCCP_ACKVEC_STATE_NOT_RECEIVED && + !seqp->ccid2s_acked) { + if (state == + DCCP_ACKVEC_STATE_ECN_MARKED) { + loss = 1; + } + else { + ccid2_new_ack(sk, seqp, + &maxincr); + } + + seqp->ccid2s_acked = 1; + ccid2_pr_debug("Got ack for %llu\n", + seqp->ccid2s_seq); + ccid2_hc_tx_dec_pipe(hctx); + } + if (seqp == hctx->ccid2hctx_seqt) { + done = 1; + break; + } + seqp = seqp->ccid2s_next; + } + if (done) + break; + + + dccp_set_seqno(&ackno, ackno_end_rl - 1); + vector++; + } + if (done) + break; + } + + /* The state about what is acked should be correct now + * Check for NUMDUPACK + */ + seqp = hctx->ccid2hctx_seqh->ccid2s_prev; + done = 0; + while (1) { + if (seqp->ccid2s_acked) { + done++; + if (done == hctx->ccid2hctx_numdupack) { + break; + } + } + if (seqp == hctx->ccid2hctx_seqt) { + break; + } + seqp = seqp->ccid2s_prev; + } + + /* If there are at least 3 acknowledgements, anything unacknowledged + * below the last sequence number is considered lost + */ + if (done == hctx->ccid2hctx_numdupack) { + struct ccid2_seq *last_acked = seqp; + + /* check for lost packets */ + while (1) { + if (!seqp->ccid2s_acked) { + loss = 1; + ccid2_hc_tx_dec_pipe(hctx); + } + if (seqp == hctx->ccid2hctx_seqt) + break; + seqp = seqp->ccid2s_prev; + } + + hctx->ccid2hctx_seqt = last_acked; + } + + /* trim acked packets in tail */ + while (hctx->ccid2hctx_seqt != hctx->ccid2hctx_seqh) { + if (!hctx->ccid2hctx_seqt->ccid2s_acked) + break; + + hctx->ccid2hctx_seqt = hctx->ccid2hctx_seqt->ccid2s_next; + } + + if (loss) { + /* XXX do bit shifts guarantee a 0 as the new bit? */ + ccid2_change_cwnd(sk, hctx->ccid2hctx_cwnd >> 1); + hctx->ccid2hctx_ssthresh = hctx->ccid2hctx_cwnd; + if (hctx->ccid2hctx_ssthresh < 2) + hctx->ccid2hctx_ssthresh = 2; + } + + ccid2_hc_tx_check_sanity(hctx); +} + +static int ccid2_hc_tx_init(struct sock *sk) +{ + struct dccp_sock *dp = dccp_sk(sk); + struct ccid2_hc_tx_sock *hctx; + int seqcount = ccid2_seq_len; + int i; + + dp->dccps_hc_tx_ccid_private = kzalloc(sizeof(*hctx), gfp_any()); + if (dp->dccps_hc_tx_ccid_private == NULL) + return -ENOMEM; + + hctx = ccid2_hc_tx_sk(sk); + + /* XXX init variables with proper values */ + hctx->ccid2hctx_cwnd = 1; + hctx->ccid2hctx_ssthresh = 10; + hctx->ccid2hctx_numdupack = 3; + + /* XXX init ~ to window size... */ + hctx->ccid2hctx_seqbuf = kmalloc(sizeof(*hctx->ccid2hctx_seqbuf) * + seqcount, gfp_any()); + if (hctx->ccid2hctx_seqbuf == NULL) { + kfree(dp->dccps_hc_tx_ccid_private); + dp->dccps_hc_tx_ccid_private = NULL; + return -ENOMEM; + } + for (i = 0; i < (seqcount - 1); i++) { + hctx->ccid2hctx_seqbuf[i].ccid2s_next = + &hctx->ccid2hctx_seqbuf[i + 1]; + hctx->ccid2hctx_seqbuf[i + 1].ccid2s_prev = + &hctx->ccid2hctx_seqbuf[i]; + } + hctx->ccid2hctx_seqbuf[seqcount - 1].ccid2s_next = + hctx->ccid2hctx_seqbuf; + hctx->ccid2hctx_seqbuf->ccid2s_prev = + &hctx->ccid2hctx_seqbuf[seqcount - 1]; + + hctx->ccid2hctx_seqh = hctx->ccid2hctx_seqbuf; + hctx->ccid2hctx_seqt = hctx->ccid2hctx_seqh; + hctx->ccid2hctx_sent = 0; + hctx->ccid2hctx_rto = 3 * HZ; + hctx->ccid2hctx_srtt = -1; + hctx->ccid2hctx_rttvar = -1; + hctx->ccid2hctx_lastrtt = 0; + hctx->ccid2hctx_rpdupack = -1; + + hctx->ccid2hctx_rtotimer.function = &ccid2_hc_tx_rto_expire; + hctx->ccid2hctx_rtotimer.data = (unsigned long)sk; + init_timer(&hctx->ccid2hctx_rtotimer); + + ccid2_hc_tx_check_sanity(hctx); + return 0; +} + +static void ccid2_hc_tx_exit(struct sock *sk) +{ + struct dccp_sock *dp = dccp_sk(sk); + struct ccid2_hc_tx_sock *hctx = dp->dccps_hc_tx_ccid_private; + + ccid2_hc_tx_kill_rto_timer(hctx); + + kfree(hctx->ccid2hctx_seqbuf); + + kfree(dp->dccps_hc_tx_ccid_private); + dp->dccps_hc_tx_ccid_private = NULL; +} + +static void ccid2_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) +{ + const struct dccp_sock *dp = dccp_sk(sk); + struct ccid2_hc_rx_sock *hcrx = ccid2_hc_rx_sk(sk); + + switch (DCCP_SKB_CB(skb)->dccpd_type) { + case DCCP_PKT_DATA: + case DCCP_PKT_DATAACK: + hcrx->ccid2hcrx_data++; + if (hcrx->ccid2hcrx_data >= dp->dccps_r_ack_ratio) { + dccp_send_ack(sk); + hcrx->ccid2hcrx_data = 0; + } + break; + } +} + +static int ccid2_hc_rx_init(struct sock *sk) +{ + struct dccp_sock *dp = dccp_sk(sk); + dp->dccps_hc_rx_ccid_private = kzalloc(sizeof(struct ccid2_hc_rx_sock), + gfp_any()); + return dp->dccps_hc_rx_ccid_private == NULL ? -ENOMEM : 0; +} + +static void ccid2_hc_rx_exit(struct sock *sk) +{ + struct dccp_sock *dp = dccp_sk(sk); + + kfree(dp->dccps_hc_rx_ccid_private); + dp->dccps_hc_rx_ccid_private = NULL; +} + +static struct ccid ccid2 = { + .ccid_id = 2, + .ccid_name = "ccid2", + .ccid_owner = THIS_MODULE, + .ccid_hc_tx_init = ccid2_hc_tx_init, + .ccid_hc_tx_exit = ccid2_hc_tx_exit, + .ccid_hc_tx_send_packet = ccid2_hc_tx_send_packet, + .ccid_hc_tx_packet_sent = ccid2_hc_tx_packet_sent, + .ccid_hc_tx_packet_recv = ccid2_hc_tx_packet_recv, + .ccid_hc_rx_init = ccid2_hc_rx_init, + .ccid_hc_rx_exit = ccid2_hc_rx_exit, + .ccid_hc_rx_packet_recv = ccid2_hc_rx_packet_recv, +}; + +module_param(ccid2_debug, int, 0444); +MODULE_PARM_DESC(ccid2_debug, "Enable debug messages"); + +static __init int ccid2_module_init(void) +{ + return ccid_register(&ccid2); +} +module_init(ccid2_module_init); + +static __exit void ccid2_module_exit(void) +{ + ccid_unregister(&ccid2); +} +module_exit(ccid2_module_exit); + +MODULE_AUTHOR("Andrea Bittau "); +MODULE_DESCRIPTION("DCCP TCP CCID2 CCID"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("net-dccp-ccid-2"); diff --git a/net/dccp/ccids/ccid2.h b/net/dccp/ccids/ccid2.h new file mode 100644 index 000000000000..0b08c90955a9 --- /dev/null +++ b/net/dccp/ccids/ccid2.h @@ -0,0 +1,69 @@ +/* + * net/dccp/ccids/ccid2.h + * + * Copyright (c) 2005 Andrea Bittau + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ +#ifndef _DCCP_CCID2_H_ +#define _DCCP_CCID2_H_ + +struct ccid2_seq { + u64 ccid2s_seq; + unsigned long ccid2s_sent; + int ccid2s_acked; + struct ccid2_seq *ccid2s_prev; + struct ccid2_seq *ccid2s_next; +}; + +/** struct ccid2_hc_tx_sock - CCID2 TX half connection + * + * @ccid2hctx_ssacks - ACKs recv in slow start + * @ccid2hctx_acks - ACKS recv in AI phase + * @ccid2hctx_sent - packets sent in this window + * @ccid2hctx_lastrtt -time RTT was last measured + * @ccid2hctx_arsent - packets sent [ack ratio] + * @ccid2hctx_ackloss - ack was lost in this win + * @ccid2hctx_rpseq - last consecutive seqno + * @ccid2hctx_rpdupack - dupacks since rpseq +*/ +struct ccid2_hc_tx_sock { + int ccid2hctx_cwnd; + int ccid2hctx_ssacks; + int ccid2hctx_acks; + int ccid2hctx_ssthresh; + int ccid2hctx_pipe; + int ccid2hctx_numdupack; + struct ccid2_seq *ccid2hctx_seqbuf; + struct ccid2_seq *ccid2hctx_seqh; + struct ccid2_seq *ccid2hctx_seqt; + long ccid2hctx_rto; + long ccid2hctx_srtt; + long ccid2hctx_rttvar; + int ccid2hctx_sent; + unsigned long ccid2hctx_lastrtt; + struct timer_list ccid2hctx_rtotimer; + unsigned long ccid2hctx_arsent; + int ccid2hctx_ackloss; + u64 ccid2hctx_rpseq; + int ccid2hctx_rpdupack; + int ccid2hctx_sendwait; +}; + +struct ccid2_hc_rx_sock { + int ccid2hcrx_data; +}; + +#endif /* _DCCP_CCID2_H_ */ diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 2ab6f0e6cd62..38321ad81875 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -1081,6 +1081,7 @@ int dccp_v4_init_sock(struct sock *sk) dp->dccps_mss_cache = 536; dp->dccps_role = DCCP_ROLE_UNDEFINED; dp->dccps_service = DCCP_SERVICE_INVALID_VALUE; + dp->dccps_l_ack_ratio = dp->dccps_r_ack_ratio = 1; return 0; } -- cgit v1.2.3-59-g8ed1b