From 686106ff5ed2cbcd3fc12a22e53c880cf0943eff Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 8 Dec 2006 13:59:52 +1100 Subject: [PATCH 01/32] [CRYPTO] sha512: Fix sha384 block size The SHA384 block size should be 128 bytes, not 96 bytes. This was spotted by Andrew Donofrio. Fortunately the block size isn't actually used anywhere so this typo has had no real impact. Signed-off-by: Herbert Xu --- crypto/sha512.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/crypto/sha512.c b/crypto/sha512.c index 2dfe7f170b48..15eab9db9be4 100644 --- a/crypto/sha512.c +++ b/crypto/sha512.c @@ -24,7 +24,7 @@ #define SHA384_DIGEST_SIZE 48 #define SHA512_DIGEST_SIZE 64 -#define SHA384_HMAC_BLOCK_SIZE 96 +#define SHA384_HMAC_BLOCK_SIZE 128 #define SHA512_HMAC_BLOCK_SIZE 128 struct sha512_ctx { From ed407717852895e04f9e088e09a4f27bb56712b7 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 9 Dec 2006 13:58:42 -0800 Subject: [PATCH 02/32] [NET]: Fix WAN routers kconfig dependency. Currently WAN router drivers can be built in-kernel while the register/unregister_wan_device interfaces are built as modules. This causes: drivers/built-in.o: In function `cycx_init': cycx_main.c:(.init.text+0x5c4b): undefined reference to `register_wan_device' drivers/built-in.o: In function `cycx_exit': cycx_main.c:(.exit.text+0x560): undefined reference to `unregister_wan_device' make: *** [.tmp_vmlinux1] Error 1 The problem is caused by tristate -> bool conversion (y or m => y), so convert WAN_ROUTER_DRIVERS to a tristate so that the correct dependency is preserved. Signed-off-by: Randy Dunlap Signed-off-by: David S. Miller --- drivers/net/wan/Kconfig | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig index d5ab9cf13257..21f76f51c95e 100644 --- a/drivers/net/wan/Kconfig +++ b/drivers/net/wan/Kconfig @@ -382,7 +382,7 @@ config SDLA # Wan router core. config WAN_ROUTER_DRIVERS - bool "WAN router drivers" + tristate "WAN router drivers" depends on WAN && WAN_ROUTER ---help--- Connect LAN to WAN via Linux box. @@ -393,7 +393,8 @@ config WAN_ROUTER_DRIVERS . Note that the answer to this question won't directly affect the - kernel: saying N will just cause the configurator to skip all + kernel except for how subordinate drivers may be built: + saying N will just cause the configurator to skip all the questions about WAN router drivers. If unsure, say N. From 2653a47b8953f500be13207bdcbcfe803bc40c88 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sat, 9 Dec 2006 13:59:25 -0800 Subject: [PATCH 03/32] [ATM]: Ignore generated file pca200e_ecd.bin2 Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- drivers/atm/.gitignore | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/atm/.gitignore b/drivers/atm/.gitignore index a165b7167714..fc0ae5eb05d8 100644 --- a/drivers/atm/.gitignore +++ b/drivers/atm/.gitignore @@ -2,4 +2,4 @@ fore200e_mkfirm fore200e_pca_fw.c pca200e.bin - +pca200e_ecd.bin2 From e37b8d931936f88e4b2af66304eb44e69510b86c Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sat, 9 Dec 2006 14:01:49 -0800 Subject: [PATCH 04/32] [NETPOLL]: Make sure TX lock is taken with BH disabled. Acked-by: Herbert Xu Signed-off-by: David S. Miller --- net/core/netpoll.c | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 8a271285f2f3..156d7fd81fec 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -242,22 +242,26 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) /* don't get messages out of order, and no recursion */ if (skb_queue_len(&npinfo->txq) == 0 && - npinfo->poll_owner != smp_processor_id() && - netif_tx_trylock(dev)) { - /* try until next clock tick */ - for (tries = jiffies_to_usecs(1)/USEC_PER_POLL; tries > 0; --tries) { - if (!netif_queue_stopped(dev)) - status = dev->hard_start_xmit(skb, dev); + npinfo->poll_owner != smp_processor_id()) { + local_bh_disable(); /* Where's netif_tx_trylock_bh()? */ + if (netif_tx_trylock(dev)) { + /* try until next clock tick */ + for (tries = jiffies_to_usecs(1)/USEC_PER_POLL; + tries > 0; --tries) { + if (!netif_queue_stopped(dev)) + status = dev->hard_start_xmit(skb, dev); - if (status == NETDEV_TX_OK) - break; + if (status == NETDEV_TX_OK) + break; - /* tickle device maybe there is some cleanup */ - netpoll_poll(np); + /* tickle device maybe there is some cleanup */ + netpoll_poll(np); - udelay(USEC_PER_POLL); + udelay(USEC_PER_POLL); + } + netif_tx_unlock(dev); } - netif_tx_unlock(dev); + local_bh_enable(); } if (status != NETDEV_TX_OK) { From 5cc3741d6cc9f07d8ddd9c45cb5088460ce3364f Mon Sep 17 00:00:00 2001 From: Ian McDonald Date: Sat, 9 Dec 2006 23:56:09 -0200 Subject: [PATCH 05/32] [DCCP]: Remove timeo from output.c It simplifies waiting for the CCID module to signal that a packet is ready to be sent. Other simplifications flow on from this such as removing constants. As a result of this EAGAIN is not returned any more by dccp_wait_for_ccid (which would otherwise lead to unnecessarily discarding the packet in dccp_write_xmit). Signed-off-by: Ian McDonald Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/dccp.h | 2 -- net/dccp/output.c | 26 ++++++-------------------- 2 files changed, 6 insertions(+), 22 deletions(-) diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index 68886986c8e4..44829d636c93 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -80,8 +80,6 @@ extern void dccp_time_wait(struct sock *sk, int state, int timeo); #define DCCP_RTO_MAX ((unsigned)(120 * HZ)) /* FIXME: using TCP value */ -#define DCCP_XMIT_TIMEO 30000 /* Time/msecs for blocking transmit per packet */ - /* sysctl variables for DCCP */ extern int sysctl_dccp_request_retries; extern int sysctl_dccp_retries1; diff --git a/net/dccp/output.c b/net/dccp/output.c index 400c30b6fcae..29a8e8508462 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -175,14 +175,12 @@ void dccp_write_space(struct sock *sk) /** * dccp_wait_for_ccid - Wait for ccid to tell us we can send a packet * @sk: socket to wait for - * @timeo: for how long */ -static int dccp_wait_for_ccid(struct sock *sk, struct sk_buff *skb, - long *timeo) +static int dccp_wait_for_ccid(struct sock *sk, struct sk_buff *skb) { struct dccp_sock *dp = dccp_sk(sk); DEFINE_WAIT(wait); - long delay; + unsigned long delay; int rc; while (1) { @@ -190,8 +188,6 @@ static int dccp_wait_for_ccid(struct sock *sk, struct sk_buff *skb, if (sk->sk_err) goto do_error; - if (!*timeo) - goto do_nonblock; if (signal_pending(current)) goto do_interrupted; @@ -199,12 +195,9 @@ static int dccp_wait_for_ccid(struct sock *sk, struct sk_buff *skb, if (rc <= 0) break; delay = msecs_to_jiffies(rc); - if (delay > *timeo || delay < 0) - goto do_nonblock; - sk->sk_write_pending++; release_sock(sk); - *timeo -= schedule_timeout(delay); + schedule_timeout(delay); lock_sock(sk); sk->sk_write_pending--; } @@ -215,11 +208,8 @@ out: do_error: rc = -EPIPE; goto out; -do_nonblock: - rc = -EAGAIN; - goto out; do_interrupted: - rc = sock_intr_errno(*timeo); + rc = -EINTR; goto out; } @@ -240,8 +230,6 @@ void dccp_write_xmit(struct sock *sk, int block) { struct dccp_sock *dp = dccp_sk(sk); struct sk_buff *skb; - long timeo = DCCP_XMIT_TIMEO; /* If a packet is taking longer than - this we have other issues */ while ((skb = skb_peek(&sk->sk_write_queue))) { int err = ccid_hc_tx_send_packet(dp->dccps_hc_tx_ccid, sk, skb); @@ -251,10 +239,8 @@ void dccp_write_xmit(struct sock *sk, int block) sk_reset_timer(sk, &dp->dccps_xmit_timer, msecs_to_jiffies(err)+jiffies); break; - } else { - err = dccp_wait_for_ccid(sk, skb, &timeo); - timeo = DCCP_XMIT_TIMEO; - } + } else + err = dccp_wait_for_ccid(sk, skb); if (err) DCCP_BUG("err=%d after dccp_wait_for_ccid", err); } From 5fce9a2da8a3b4ed088d9c0ac7c938a638044fdb Mon Sep 17 00:00:00 2001 From: Ian McDonald Date: Sat, 9 Dec 2006 23:58:10 -0200 Subject: [PATCH 06/32] [DCCP]: Update documentation As Eddie Kohler points out the RFC is Proposed Standard not experimental. Also removed documentation about deprecated socket option. Signed-off-by: Ian McDonald Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- Documentation/networking/dccp.txt | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index dda15886bcb5..387482e46c47 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -19,7 +19,8 @@ for real time and multimedia traffic. It has a base protocol and pluggable congestion control IDs (CCIDs). -It is at experimental RFC status and the homepage for DCCP as a protocol is at: +It is at proposed standard RFC status and the homepage for DCCP as a protocol +is at: http://www.read.cs.ucla.edu/dccp/ Missing features @@ -34,9 +35,6 @@ The known bugs are at: Socket options ============== -DCCP_SOCKOPT_PACKET_SIZE is used for CCID3 to set default packet size for -calculations. - 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, the socket will fall back to 0 (which means that no meaningful service code From 45393a66a2c30fc8ed86e6c79a63a3cd5231c69f Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sat, 9 Dec 2006 23:59:14 -0200 Subject: [PATCH 07/32] [DCCP] ccid3: Check against too large p This patch follows a suggestion by Ian McDonald and ensures that in the current code the value of p can not exceed 100%. Such a value is illegal and would consequently cause a bug condition in tfrc_calc_x(). The receiver case is also tested, and a warning message is added. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 66a27b9688ca..f1b745ee9cb6 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -444,9 +444,9 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) /* Update loss event rate */ pinv = opt_recv->ccid3or_loss_event_rate; - if (pinv == ~0U || pinv == 0) + if (pinv == ~0U || pinv == 0) /* see RFC 4342, 8.5 */ hctx->ccid3hctx_p = 0; - else + else /* can not exceed 100% */ hctx->ccid3hctx_p = 1000000 / pinv; dccp_timestamp(sk, &now); @@ -733,10 +733,15 @@ static void ccid3_hc_rx_send_feedback(struct sock *sk) /* Convert to multiples of 10us */ hcrx->ccid3hcrx_elapsed_time = timeval_delta(&now, &packet->dccphrx_tstamp) / 10; + if (hcrx->ccid3hcrx_p == 0) - hcrx->ccid3hcrx_pinv = ~0; - else + hcrx->ccid3hcrx_pinv = ~0U; /* see RFC 4342, 8.5 */ + else if (hcrx->ccid3hcrx_p > 1000000) { + DCCP_WARN("p (%u) > 100%%\n", hcrx->ccid3hcrx_p); + hcrx->ccid3hcrx_pinv = 1; /* use 100% in this case */ + } else hcrx->ccid3hcrx_pinv = 1000000 / hcrx->ccid3hcrx_p; + dp->dccps_hc_rx_insert_options = 1; dccp_send_ack(sk); } From ff58629824c068e2a75402b5b83f12af0b36d394 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:00:14 -0200 Subject: [PATCH 08/32] [DCCP] ccid3: Two optimisations for sending rate recomputation This performs two optimisations for the recomputation of the sending rate. 1) Currently the target sending rate X_calc is recalculated whenever a) the nofeedback timer expires, or b) a feedback packet is received. In the (a) case, recomputing X_calc is redundant, since * the parameters p and RTT do not change in between the reception of feedback packets; * the parameter X_recv is either modified from received feedback or via the nofeedback timer; * a test (`p == 0') in the nofeedback timer avoids using a stale/undefined value of X_calc if p was previously 0. 2) The nofeedback timer now only recomputes a timestamp when p == 0. This is according to step (4) of [RFC 3448, 4.3] and avoids unnecessarily determining a timestamp. A debug statement about not updating X is also removed - it helps very little in debugging and just clutters the logs. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index f1b745ee9cb6..92e893ee77c2 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -138,9 +138,6 @@ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) const __u32 old_x = hctx->ccid3hctx_x; if (hctx->ccid3hctx_p > 0) { - hctx->ccid3hctx_x_calc = tfrc_calc_x(hctx->ccid3hctx_s, - hctx->ccid3hctx_rtt, - hctx->ccid3hctx_p); hctx->ccid3hctx_x = max_t(u32, min(hctx->ccid3hctx_x_calc, hctx->ccid3hctx_x_recv * 2), hctx->ccid3hctx_s / TFRC_T_MBI); @@ -152,8 +149,7 @@ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) usecs_div(hctx->ccid3hctx_s, hctx->ccid3hctx_rtt) ); hctx->ccid3hctx_t_ld = *now; - } else - ccid3_pr_debug("Not changing X\n"); + } if (hctx->ccid3hctx_x != old_x) ccid3_update_send_time(hctx); @@ -223,9 +219,11 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) ccid3_pr_debug("%s, sk=%p, state=%s, not idle\n", dccp_role(sk), sk, ccid3_tx_state_name(hctx->ccid3hctx_state)); - /* Halve sending rate */ - /* If (p == 0 || X_calc > 2 * X_recv) + /* + * Modify the cached value of X_recv [RFC 3448, 4.4] + * + * If (p == 0 || X_calc > 2 * X_recv) * X_recv = max(X_recv / 2, s / (2 * t_mbi)); * Else * X_recv = X_calc / 4; @@ -233,14 +231,15 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) BUG_ON(hctx->ccid3hctx_p && !hctx->ccid3hctx_x_calc); if (hctx->ccid3hctx_p == 0 || - hctx->ccid3hctx_x_calc > 2 * hctx->ccid3hctx_x_recv) + hctx->ccid3hctx_x_calc > 2 * hctx->ccid3hctx_x_recv) { hctx->ccid3hctx_x_recv = max_t(u32, hctx->ccid3hctx_x_recv / 2, hctx->ccid3hctx_s / (2 * TFRC_T_MBI)); - else + if (hctx->ccid3hctx_p == 0) + dccp_timestamp(sk, &now); + } else hctx->ccid3hctx_x_recv = hctx->ccid3hctx_x_calc / 4; - /* Update sending rate */ - dccp_timestamp(sk, &now); + /* Now recalculate X [RFC 3448, 4.3, step (4)] */ ccid3_hc_tx_update_x(sk, &now); } /* @@ -496,6 +495,12 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) hctx->ccid3hctx_rtt = (9 * hctx->ccid3hctx_rtt + (u32)r_sample ) / 10; + /* Update sending rate (step 4 of [RFC 3448, 4.3]) */ + if (hctx->ccid3hctx_p > 0) + hctx->ccid3hctx_x_calc = + tfrc_calc_x(hctx->ccid3hctx_s, + hctx->ccid3hctx_rtt, + hctx->ccid3hctx_p); ccid3_hc_tx_update_x(sk, &now); ccid3_pr_debug("%s(%p), RTT=%uus (sample=%ldus), s=%u, " From 179ebc9f92da88e15ea86d7d27308c92712d8ee9 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:01:22 -0200 Subject: [PATCH 09/32] [DCCP] ccid3: Fix two bugs in sending rate computation This fixes 1) a bug in the recomputation of the sending rate by the nofeedback timer when no feedback at all has so far been sent by the receiver: min_t was used instead of max_t, which is wrong (cf. RFC 3448, p. 10); 2) an error in the computation of larger initial windows: instead of min(... max()) (cf. RFC 4342, 5.), the code had used max(... max()). Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 92e893ee77c2..c54663f21fdd 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -193,7 +193,7 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) switch (hctx->ccid3hctx_state) { case TFRC_SSTATE_NO_FBACK: /* RFC 3448, 4.4: Halve send rate directly */ - hctx->ccid3hctx_x = min_t(u32, hctx->ccid3hctx_x / 2, + hctx->ccid3hctx_x = max_t(u32, hctx->ccid3hctx_x / 2, hctx->ccid3hctx_s / TFRC_T_MBI); ccid3_pr_debug("%s, sk=%p, state=%s, updated tx rate to %d " @@ -477,7 +477,7 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) if (hctx->ccid3hctx_state == TFRC_SSTATE_NO_FBACK) { /* Use Larger Initial Windows [RFC 4342, sec. 5] * We deviate in that we use `s' instead of `MSS'. */ - u16 w_init = max( 4 * hctx->ccid3hctx_s, + u16 w_init = min( 4 * hctx->ccid3hctx_s, max(2 * hctx->ccid3hctx_s, 4380)); hctx->ccid3hctx_rtt = r_sample; hctx->ccid3hctx_x = usecs_div(w_init, r_sample); From 1a21e49a8d60f588c1276f765198b14d5688a778 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:02:12 -0200 Subject: [PATCH 10/32] [DCCP] ccid3: Finer-grained resolution of sending rates This patch * resolves a bug where packets smaller than 32/64 bytes resulted in sending rates of 0 * supports all sending rates from 1/64 bytes/second up to 4Gbyte/second * simplifies the present overflow problems in calculations Current sending rate X and the cached value X_recv of the receiver-estimated sending rate are both scaled by 64 (2^6) in order to * cope with low sending rates (minimally 1 byte/second) * allow upgrading to use a packets-per-second implementation of CCID 3 * avoid calculation errors due to integer arithmetic cut-off The patch implements a revised strategy from http://www.mail-archive.com/dccp@vger.kernel.org/msg01040.html The only difference with regard to that strategy is that t_ipi is already used in the calculation of the nofeedback timeout, which saves one division. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/tfrc.h | 8 +++-- net/dccp/ccids/ccid3.c | 76 +++++++++++++++++++++++++----------------- net/dccp/ccids/ccid3.h | 33 ++++++++++++++---- 3 files changed, 78 insertions(+), 39 deletions(-) diff --git a/include/linux/tfrc.h b/include/linux/tfrc.h index 31a9b25276fe..8a8462b4a4dd 100644 --- a/include/linux/tfrc.h +++ b/include/linux/tfrc.h @@ -37,10 +37,14 @@ struct tfrc_rx_info { * @tfrctx_p: current loss event rate (5.4) * @tfrctx_rto: estimate of RTO, equals 4*RTT (4.3) * @tfrctx_ipi: inter-packet interval (4.6) + * + * Note: X and X_recv are both maintained in units of 64 * bytes/second. This + * enables a finer resolution of sending rates and avoids problems with + * integer arithmetic; u32 is not sufficient as scaling consumes 6 bits. */ struct tfrc_tx_info { - __u32 tfrctx_x; - __u32 tfrctx_x_recv; + __u64 tfrctx_x; + __u64 tfrctx_x_recv; __u32 tfrctx_x_calc; __u32 tfrctx_rtt; __u32 tfrctx_p; diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index c54663f21fdd..aa355d4cfc8a 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -108,8 +108,9 @@ static inline void ccid3_update_send_time(struct ccid3_hc_tx_sock *hctx) { timeval_sub_usecs(&hctx->ccid3hctx_t_nom, hctx->ccid3hctx_t_ipi); - /* Calculate new t_ipi (inter packet interval) by t_ipi = s / X_inst */ - hctx->ccid3hctx_t_ipi = usecs_div(hctx->ccid3hctx_s, hctx->ccid3hctx_x); + /* Calculate new t_ipi = s / X_inst (X_inst is in 64 * bytes/second) */ + hctx->ccid3hctx_t_ipi = scaled_div(hctx->ccid3hctx_s, + hctx->ccid3hctx_x >> 6); /* Update nominal send time with regard to the new t_ipi */ timeval_add_usecs(&hctx->ccid3hctx_t_nom, hctx->ccid3hctx_t_ipi); @@ -128,26 +129,33 @@ static inline void ccid3_update_send_time(struct ccid3_hc_tx_sock *hctx) * X = max(min(2 * X, 2 * X_recv), s / R); * tld = now; * + * Note: X and X_recv are both stored in units of 64 * bytes/second, to support + * fine-grained resolution of sending rates. This requires scaling by 2^6 + * throughout the code. Only X_calc is unscaled (in bytes/second). + * * If X has changed, we also update the scheduled send time t_now, * the inter-packet interval t_ipi, and the delta value. - */ + */ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) { struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); - const __u32 old_x = hctx->ccid3hctx_x; + const __u64 old_x = hctx->ccid3hctx_x; if (hctx->ccid3hctx_p > 0) { - hctx->ccid3hctx_x = max_t(u32, min(hctx->ccid3hctx_x_calc, - hctx->ccid3hctx_x_recv * 2), - hctx->ccid3hctx_s / TFRC_T_MBI); + + hctx->ccid3hctx_x = min_t(u64, hctx->ccid3hctx_x_calc << 6, + hctx->ccid3hctx_x_recv * 2 ); + hctx->ccid3hctx_x = max_t(u64, hctx->ccid3hctx_x, + (hctx->ccid3hctx_s << 6)/TFRC_T_MBI); } else if (timeval_delta(now, &hctx->ccid3hctx_t_ld) >= hctx->ccid3hctx_rtt) { - hctx->ccid3hctx_x = max(min(hctx->ccid3hctx_x_recv, - hctx->ccid3hctx_x ) * 2, - usecs_div(hctx->ccid3hctx_s, - hctx->ccid3hctx_rtt) ); + + hctx->ccid3hctx_x = max(2 * min(hctx->ccid3hctx_x, + hctx->ccid3hctx_x_recv), + scaled_div(hctx->ccid3hctx_s << 6, + hctx->ccid3hctx_rtt )); hctx->ccid3hctx_t_ld = *now; } @@ -194,13 +202,13 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) case TFRC_SSTATE_NO_FBACK: /* RFC 3448, 4.4: Halve send rate directly */ hctx->ccid3hctx_x = max_t(u32, hctx->ccid3hctx_x / 2, - hctx->ccid3hctx_s / TFRC_T_MBI); + (hctx->ccid3hctx_s << 6)/TFRC_T_MBI); - ccid3_pr_debug("%s, sk=%p, state=%s, updated tx rate to %d " + ccid3_pr_debug("%s, sk=%p, state=%s, updated tx rate to %u " "bytes/s\n", dccp_role(sk), sk, ccid3_tx_state_name(hctx->ccid3hctx_state), - hctx->ccid3hctx_x); + (unsigned)(hctx->ccid3hctx_x >> 6)); /* The value of R is still undefined and so we can not recompute * the timout value. Keep initial value as per [RFC 4342, 5]. */ t_nfb = TFRC_INITIAL_TIMEOUT; @@ -209,11 +217,11 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) case TFRC_SSTATE_FBACK: /* * Check if IDLE since last timeout and recv rate is less than - * 4 packets per RTT + * 4 packets (in units of 64*bytes/sec) per RTT */ if (!hctx->ccid3hctx_idle || - (hctx->ccid3hctx_x_recv >= - 4 * usecs_div(hctx->ccid3hctx_s, hctx->ccid3hctx_rtt))) { + (hctx->ccid3hctx_x_recv >= 4 * + scaled_div(hctx->ccid3hctx_s << 6, hctx->ccid3hctx_rtt))) { struct timeval now; ccid3_pr_debug("%s, sk=%p, state=%s, not idle\n", @@ -227,17 +235,23 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) * X_recv = max(X_recv / 2, s / (2 * t_mbi)); * Else * X_recv = X_calc / 4; + * + * Note that X_recv is scaled by 2^6 while X_calc is not */ BUG_ON(hctx->ccid3hctx_p && !hctx->ccid3hctx_x_calc); if (hctx->ccid3hctx_p == 0 || - hctx->ccid3hctx_x_calc > 2 * hctx->ccid3hctx_x_recv) { - hctx->ccid3hctx_x_recv = max_t(u32, hctx->ccid3hctx_x_recv / 2, - hctx->ccid3hctx_s / (2 * TFRC_T_MBI)); + hctx->ccid3hctx_x_calc > (hctx->ccid3hctx_x_recv >> 5)) { + + hctx->ccid3hctx_x_recv = + max_t(u64, hctx->ccid3hctx_x_recv / 2, + (hctx->ccid3hctx_s << 6) / + (2*TFRC_T_MBI)); + if (hctx->ccid3hctx_p == 0) dccp_timestamp(sk, &now); } else - hctx->ccid3hctx_x_recv = hctx->ccid3hctx_x_calc / 4; + hctx->ccid3hctx_x_recv = hctx->ccid3hctx_x_calc << 4; /* Now recalculate X [RFC 3448, 4.3, step (4)] */ ccid3_hc_tx_update_x(sk, &now); @@ -315,9 +329,9 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) hctx->ccid3hctx_t_last_win_count = now; ccid3_hc_tx_set_state(sk, TFRC_SSTATE_NO_FBACK); - /* Set initial sending rate to 1 packet per second */ + /* Set initial sending rate X/s to 1pps (X is scaled by 2^6) */ ccid3_hc_tx_update_s(hctx, skb->len); - hctx->ccid3hctx_x = hctx->ccid3hctx_s; + hctx->ccid3hctx_x = hctx->ccid3hctx_s << 6; /* First timeout, according to [RFC 3448, 4.2], is 1 second */ hctx->ccid3hctx_t_ipi = USEC_PER_SEC; @@ -438,8 +452,8 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) return; } - /* Update receive rate */ - hctx->ccid3hctx_x_recv = opt_recv->ccid3or_receive_rate; + /* Update receive rate in units of 64 * bytes/second */ + hctx->ccid3hctx_x_recv = opt_recv->ccid3or_receive_rate << 6; /* Update loss event rate */ pinv = opt_recv->ccid3or_loss_event_rate; @@ -475,12 +489,14 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) * q is a constant, RFC 3448 recomments 0.9 */ if (hctx->ccid3hctx_state == TFRC_SSTATE_NO_FBACK) { - /* Use Larger Initial Windows [RFC 4342, sec. 5] - * We deviate in that we use `s' instead of `MSS'. */ + /* + * Larger Initial Windows [RFC 4342, sec. 5] + * We deviate in that we use `s' instead of `MSS'. + */ u16 w_init = min( 4 * hctx->ccid3hctx_s, max(2 * hctx->ccid3hctx_s, 4380)); hctx->ccid3hctx_rtt = r_sample; - hctx->ccid3hctx_x = usecs_div(w_init, r_sample); + hctx->ccid3hctx_x = scaled_div(w_init<< 6, r_sample); hctx->ccid3hctx_t_ld = now; ccid3_update_send_time(hctx); @@ -488,7 +504,7 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) ccid3_pr_debug("%s(%p), s=%u, w_init=%u, " "R_sample=%ldus, X=%u\n", dccp_role(sk), sk, hctx->ccid3hctx_s, w_init, r_sample, - hctx->ccid3hctx_x); + (unsigned)(hctx->ccid3hctx_x >> 6)); ccid3_hc_tx_set_state(sk, TFRC_SSTATE_FBACK); } else { @@ -508,7 +524,7 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) sk, hctx->ccid3hctx_rtt, r_sample, hctx->ccid3hctx_s, hctx->ccid3hctx_p, hctx->ccid3hctx_x_calc, - hctx->ccid3hctx_x); + (unsigned)(hctx->ccid3hctx_x >> 6)); } /* unschedule no feedback timer */ diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index 07596d704ef9..cd4fc542f73e 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -75,14 +75,14 @@ enum ccid3_hc_tx_states { /** struct ccid3_hc_tx_sock - CCID3 sender half-connection socket * - * @ccid3hctx_x - Current sending rate - * @ccid3hctx_x_recv - Receive rate - * @ccid3hctx_x_calc - Calculated send rate (RFC 3448, 3.1) + * @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 - * @ccid3hctx_t_rto - Retransmission Timeout (RFC 3448, 3.1) - * @ccid3hctx_t_ipi - Interpacket (send) interval (RFC 3448, 4.6) + * @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 @@ -91,7 +91,7 @@ enum ccid3_hc_tx_states { * @ccid3hctx_idle - Flag indicating that sender is idling * @ccid3hctx_t_ld - Time last doubled during slow start * @ccid3hctx_t_nom - Nominal send time of next packet - * @ccid3hctx_delta - Send timer delta + * @ccid3hctx_delta - Send timer delta (RFC 3448, 4.6) in usecs * @ccid3hctx_hist - Packet history * @ccid3hctx_options_received - Parsed set of retrieved options */ @@ -171,4 +171,23 @@ static inline struct ccid3_hc_rx_sock *ccid3_hc_rx_sk(const struct sock *sk) return ccid_priv(dccp_sk(sk)->dccps_hc_rx_ccid); } +static inline u64 scaled_div(u64 a, u32 b) +{ + BUG_ON(b==0); + a *= 1000000; + do_div(a, b); + return a; +} + +static inline u32 scaled_div32(u64 a, u32 b) +{ + u64 result = scaled_div(a, b); + + if (result > UINT_MAX) { + DCCP_CRIT("Overflow: a(%llu)/b(%u) > ~0U", + (unsigned long long)a, b); + return UINT_MAX; + } + return result; +} #endif /* _DCCP_CCID3_H_ */ From b9039a2a8df974d7702564318722434bb276a995 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:02:51 -0200 Subject: [PATCH 11/32] [DCCP] ccid3: Replace scaled division operations This replaces the remaining uses of usecs_div with scaled_div32, which internally uses 64bit division and produces a warning on overflow. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 27 +++------------------------ 1 file changed, 3 insertions(+), 24 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index aa355d4cfc8a..bdd13de4f422 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -41,27 +41,6 @@ #include "lib/tfrc.h" #include "ccid3.h" -/* - * Reason for maths here is to avoid 32 bit overflow when a is big. - * With this we get close to the limit. - */ -static u32 usecs_div(const u32 a, const u32 b) -{ - const u32 div = a < (UINT_MAX / (USEC_PER_SEC / 10)) ? 10 : - a < (UINT_MAX / (USEC_PER_SEC / 50)) ? 50 : - a < (UINT_MAX / (USEC_PER_SEC / 100)) ? 100 : - a < (UINT_MAX / (USEC_PER_SEC / 500)) ? 500 : - a < (UINT_MAX / (USEC_PER_SEC / 1000)) ? 1000 : - a < (UINT_MAX / (USEC_PER_SEC / 5000)) ? 5000 : - a < (UINT_MAX / (USEC_PER_SEC / 10000)) ? 10000 : - a < (UINT_MAX / (USEC_PER_SEC / 50000)) ? 50000 : - 100000; - const u32 tmp = a * (USEC_PER_SEC / div); - return (b >= 2 * div) ? tmp / (b / div) : tmp; -} - - - #ifdef CONFIG_IP_DCCP_CCID3_DEBUG static int ccid3_debug; #define ccid3_pr_debug(format, a...) DCCP_PR_DEBUG(ccid3_debug, format, ##a) @@ -731,8 +710,8 @@ static void ccid3_hc_rx_send_feedback(struct sock *sk) case TFRC_RSTATE_DATA: { const u32 delta = timeval_delta(&now, &hcrx->ccid3hcrx_tstamp_last_feedback); - hcrx->ccid3hcrx_x_recv = usecs_div(hcrx->ccid3hcrx_bytes_recv, - delta); + hcrx->ccid3hcrx_x_recv = + scaled_div32(hcrx->ccid3hcrx_bytes_recv, delta); } break; case TFRC_RSTATE_TERM: @@ -862,7 +841,7 @@ found: dccp_timestamp(sk, &tstamp); delta = timeval_delta(&tstamp, &hcrx->ccid3hcrx_tstamp_last_feedback); - x_recv = usecs_div(hcrx->ccid3hcrx_bytes_recv, delta); + x_recv = scaled_div32(hcrx->ccid3hcrx_bytes_recv, delta); if (x_recv == 0) x_recv = hcrx->ccid3hcrx_x_recv; From bfe24a6cc222d27e1491f850802fa6932232b8ef Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:03:30 -0200 Subject: [PATCH 12/32] [DCCP] ccid3: Simplify calculation for reverse lookup of p This simplifies the calculation of a value p for a given fval when the first loss interval is computed (RFC 3448, 6.3.1). It makes use of the two new functions scaled_div/scaled_div32 to provide overflow protection. Additionally, protection against divide-by-zero is extended - in this case the function will return the maximally possible value of p=100%. Background: The maximum fval, f(100%), is approximately 244, i.e. the scaled value of fval should never exceed 244E6, which fits easily into u32. The problem is the scaling by 10^6, since additionally R(TT) is in microseconds. This is resolved by breaking the division into two stages: the first stage computes fval=(s*10^6)/R, stores that into u64; the second stage computes fval = (fval*10^6)/X_recv and complains if overflow is reached for u32. This case is safe since the TFRC reverse-lookup routine then returns p=100%. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 39 ++++++++++++++++++++++----------------- 1 file changed, 22 insertions(+), 17 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index bdd13de4f422..89ef1183e48c 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -785,12 +785,12 @@ static u32 ccid3_hc_rx_calc_first_li(struct sock *sk) { struct ccid3_hc_rx_sock *hcrx = ccid3_hc_rx_sk(sk); struct dccp_rx_hist_entry *entry, *next, *tail = NULL; - u32 rtt, delta, x_recv, fval, p, tmp2; + u32 rtt, delta, x_recv, p; struct timeval tstamp = { 0, }; int interval = 0; int win_count = 0; int step = 0; - u64 tmp1; + u64 fval; list_for_each_entry_safe(entry, next, &hcrx->ccid3hcrx_hist, dccphrx_node) { @@ -834,30 +834,35 @@ found: ccid3_pr_debug("%s, sk=%p, approximated RTT to %uus\n", dccp_role(sk), sk, rtt); - if (rtt == 0) { - DCCP_WARN("RTT==0, setting to 1\n"); - rtt = 1; + /* + * Determine the length of the first loss interval via inverse lookup. + * Assume that X_recv can be computed by the throughput equation + * s + * X_recv = -------- + * R * fval + * Find some p such that f(p) = fval; return 1/p [RFC 3448, 6.3.1]. + */ + if (rtt == 0) { /* would result in divide-by-zero */ + DCCP_WARN("RTT==0, returning 1/p = 1\n"); + return 1000000; } dccp_timestamp(sk, &tstamp); delta = timeval_delta(&tstamp, &hcrx->ccid3hcrx_tstamp_last_feedback); x_recv = scaled_div32(hcrx->ccid3hcrx_bytes_recv, delta); - if (x_recv == 0) - x_recv = hcrx->ccid3hcrx_x_recv; - - tmp1 = (u64)x_recv * (u64)rtt; - do_div(tmp1,10000000); - tmp2 = (u32)tmp1; - - if (!tmp2) { - DCCP_CRIT("tmp2 = 0, x_recv = %u, rtt =%u\n", x_recv, rtt); - return ~0; + if (x_recv == 0) { /* would also trigger divide-by-zero */ + DCCP_WARN("X_recv==0\n"); + if ((x_recv = hcrx->ccid3hcrx_x_recv) == 0) { + DCCP_BUG("stored value of X_recv is zero"); + return 1000000; + } } - fval = (hcrx->ccid3hcrx_s * 100000) / tmp2; - /* do not alter order above or you will get overflow on 32 bit */ + fval = scaled_div(hcrx->ccid3hcrx_s, rtt); + fval = scaled_div32(fval, x_recv); p = tfrc_calc_x_reverse_lookup(fval); + ccid3_pr_debug("%s, sk=%p, receive rate=%u bytes/s, implied " "loss rate=%u\n", dccp_role(sk), sk, x_recv, p); From 0f9e5b573f7249b0e04a03457b55081d1f60f2bf Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:03:51 -0200 Subject: [PATCH 13/32] [DCCP]: Debug timeval operations Problem: Most target types in the CCID3 code are u32, so subtle conversion errors can occur if signed time calculations yield negative results: the original values are lost in the conversion to unsigned, calculation errors go undetected. This patch therefore * sets all critical time types from unsigned to suseconds_t * avoids comparison between signed/unsigned via type-casting * provides ample warning messages in case time calculations are negative These warning messages can be removed at a later stage when the code has undergone more testing. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 56 ++++++++++++++++++++++++------------------ net/dccp/dccp.h | 1 + 2 files changed, 33 insertions(+), 24 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 89ef1183e48c..820bc25e14dc 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -128,8 +128,8 @@ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) hctx->ccid3hctx_x = max_t(u64, hctx->ccid3hctx_x, (hctx->ccid3hctx_s << 6)/TFRC_T_MBI); - } else if (timeval_delta(now, &hctx->ccid3hctx_t_ld) >= - hctx->ccid3hctx_rtt) { + } else if (timeval_delta(now, &hctx->ccid3hctx_t_ld) - + (suseconds_t)hctx->ccid3hctx_rtt >= 0 ) { hctx->ccid3hctx_x = max(2 * min(hctx->ccid3hctx_x, hctx->ccid3hctx_x_recv), @@ -271,7 +271,7 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); struct dccp_tx_hist_entry *new_packet; struct timeval now; - long delay; + suseconds_t delay; BUG_ON(hctx == NULL); @@ -331,7 +331,7 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) * else * // send the packet in (t_nom - t_now) milliseconds. */ - if (delay - (long)hctx->ccid3hctx_delta >= 0) + if (delay - (suseconds_t)hctx->ccid3hctx_delta >= 0) return delay / 1000L; break; case TFRC_SSTATE_TERM: @@ -353,7 +353,7 @@ static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) const struct dccp_sock *dp = dccp_sk(sk); struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); struct timeval now; - unsigned long quarter_rtt; + suseconds_t quarter_rtt; struct dccp_tx_hist_entry *packet; BUG_ON(hctx == NULL); @@ -450,10 +450,8 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) r_sample = timeval_delta(&now, &packet->dccphtx_tstamp); t_elapsed = dp->dccps_options_received.dccpor_elapsed_time * 10; - if (unlikely(r_sample <= 0)) { - DCCP_WARN("WARNING: R_sample (%ld) <= 0!\n", r_sample); - r_sample = 0; - } else if (unlikely(r_sample <= t_elapsed)) + DCCP_BUG_ON(r_sample < 0); + if (unlikely(r_sample <= t_elapsed)) DCCP_WARN("WARNING: r_sample=%ldus <= t_elapsed=%ldus\n", r_sample, t_elapsed); else @@ -698,6 +696,7 @@ static void ccid3_hc_rx_send_feedback(struct sock *sk) struct dccp_sock *dp = dccp_sk(sk); struct dccp_rx_hist_entry *packet; struct timeval now; + suseconds_t delta; ccid3_pr_debug("%s, sk=%p\n", dccp_role(sk), sk); @@ -707,12 +706,12 @@ static void ccid3_hc_rx_send_feedback(struct sock *sk) case TFRC_RSTATE_NO_DATA: hcrx->ccid3hcrx_x_recv = 0; break; - case TFRC_RSTATE_DATA: { - const u32 delta = timeval_delta(&now, - &hcrx->ccid3hcrx_tstamp_last_feedback); + case TFRC_RSTATE_DATA: + delta = timeval_delta(&now, + &hcrx->ccid3hcrx_tstamp_last_feedback); + DCCP_BUG_ON(delta < 0); hcrx->ccid3hcrx_x_recv = scaled_div32(hcrx->ccid3hcrx_bytes_recv, delta); - } break; case TFRC_RSTATE_TERM: DCCP_BUG("Illegal %s state TERM, sk=%p", dccp_role(sk), sk); @@ -730,9 +729,10 @@ static void ccid3_hc_rx_send_feedback(struct sock *sk) hcrx->ccid3hcrx_ccval_last_counter = packet->dccphrx_ccval; hcrx->ccid3hcrx_bytes_recv = 0; - /* Convert to multiples of 10us */ - hcrx->ccid3hcrx_elapsed_time = - timeval_delta(&now, &packet->dccphrx_tstamp) / 10; + /* Elapsed time information [RFC 4340, 13.2] in units of 10 * usecs */ + delta = timeval_delta(&now, &packet->dccphrx_tstamp); + DCCP_BUG_ON(delta < 0); + hcrx->ccid3hcrx_elapsed_time = delta / 10; if (hcrx->ccid3hcrx_p == 0) hcrx->ccid3hcrx_pinv = ~0U; /* see RFC 4342, 8.5 */ @@ -785,7 +785,8 @@ static u32 ccid3_hc_rx_calc_first_li(struct sock *sk) { struct ccid3_hc_rx_sock *hcrx = ccid3_hc_rx_sk(sk); struct dccp_rx_hist_entry *entry, *next, *tail = NULL; - u32 rtt, delta, x_recv, p; + u32 x_recv, p; + suseconds_t rtt, delta; struct timeval tstamp = { 0, }; int interval = 0; int win_count = 0; @@ -830,8 +831,12 @@ found: DCCP_CRIT("tail is null\n"); return ~0; } - rtt = timeval_delta(&tstamp, &tail->dccphrx_tstamp) * 4 / interval; - ccid3_pr_debug("%s, sk=%p, approximated RTT to %uus\n", + + delta = timeval_delta(&tstamp, &tail->dccphrx_tstamp); + DCCP_BUG_ON(delta < 0); + + rtt = delta * 4 / interval; + ccid3_pr_debug("%s, sk=%p, approximated RTT to %ldus\n", dccp_role(sk), sk, rtt); /* @@ -849,8 +854,9 @@ found: dccp_timestamp(sk, &tstamp); delta = timeval_delta(&tstamp, &hcrx->ccid3hcrx_tstamp_last_feedback); - x_recv = scaled_div32(hcrx->ccid3hcrx_bytes_recv, delta); + DCCP_BUG_ON(delta <= 0); + x_recv = scaled_div32(hcrx->ccid3hcrx_bytes_recv, delta); if (x_recv == 0) { /* would also trigger divide-by-zero */ DCCP_WARN("X_recv==0\n"); if ((x_recv = hcrx->ccid3hcrx_x_recv) == 0) { @@ -977,7 +983,8 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) const struct dccp_options_received *opt_recv; struct dccp_rx_hist_entry *packet; struct timeval now; - u32 p_prev, rtt_prev, r_sample, t_elapsed; + u32 p_prev, rtt_prev; + suseconds_t r_sample, t_elapsed; int loss, payload_size; BUG_ON(hcrx == NULL); @@ -997,8 +1004,9 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) r_sample = timeval_usecs(&now); t_elapsed = opt_recv->dccpor_elapsed_time * 10; + DCCP_BUG_ON(r_sample < 0); if (unlikely(r_sample <= t_elapsed)) - DCCP_WARN("r_sample=%uus, t_elapsed=%uus\n", + DCCP_WARN("r_sample=%ldus, t_elapsed=%ldus\n", r_sample, t_elapsed); else r_sample -= t_elapsed; @@ -1051,8 +1059,8 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) break; dccp_timestamp(sk, &now); - if (timeval_delta(&now, &hcrx->ccid3hcrx_tstamp_last_ack) >= - hcrx->ccid3hcrx_rtt) { + if (timeval_delta(&now, &hcrx->ccid3hcrx_tstamp_last_ack) - + (suseconds_t)hcrx->ccid3hcrx_rtt >= 0) { hcrx->ccid3hcrx_tstamp_last_ack = now; ccid3_hc_rx_send_feedback(sk); } diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index 44829d636c93..a0900bf98e6b 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -432,6 +432,7 @@ static inline void timeval_sub_usecs(struct timeval *tv, tv->tv_sec--; tv->tv_usec += USEC_PER_SEC; } + DCCP_BUG_ON(tv->tv_sec < 0); } #ifdef CONFIG_SYSCTL From d63d8364cfe17fc9bb738f554f452595f76f21d2 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:04:16 -0200 Subject: [PATCH 14/32] [DCCP]: Simplify TFRC calculation In migrating towards using the newer functions scaled_div/scaled_div32 for TFRC computations mapped from floating-point onto integer arithmetic, this completes the last stage of modifications. In particular, the overflow case for computing X_calc is circumvented by * breaking the computation into two stages * the first stage, res = (s*1E6)/R, cannot overflow due to use of u64 * in the second stage, res = (res*1E6)/f, overflow on u32 is avoided due to (i) returning UINT_MAX in this case (which is logically appropriate) and (ii) issuing a warning message into the system log (since very likely there is a problem somewhere else with the parameters) Lastly, all such scaling operations are now exported into tfrc.h, since actually this form of scaled computation is specific to TFRC and not to CCID3. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.h | 19 ------------------- net/dccp/ccids/lib/tfrc.h | 23 ++++++++++++++++++++++- net/dccp/ccids/lib/tfrc_equation.c | 28 +++++++++++++--------------- 3 files changed, 35 insertions(+), 35 deletions(-) diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index cd4fc542f73e..da0ca3c0a7b3 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -171,23 +171,4 @@ static inline struct ccid3_hc_rx_sock *ccid3_hc_rx_sk(const struct sock *sk) return ccid_priv(dccp_sk(sk)->dccps_hc_rx_ccid); } -static inline u64 scaled_div(u64 a, u32 b) -{ - BUG_ON(b==0); - a *= 1000000; - do_div(a, b); - return a; -} - -static inline u32 scaled_div32(u64 a, u32 b) -{ - u64 result = scaled_div(a, b); - - if (result > UINT_MAX) { - DCCP_CRIT("Overflow: a(%llu)/b(%u) > ~0U", - (unsigned long long)a, b); - return UINT_MAX; - } - return result; -} #endif /* _DCCP_CCID3_H_ */ diff --git a/net/dccp/ccids/lib/tfrc.h b/net/dccp/ccids/lib/tfrc.h index 45f30f59ea2a..faf5f7e219e3 100644 --- a/net/dccp/ccids/lib/tfrc.h +++ b/net/dccp/ccids/lib/tfrc.h @@ -13,8 +13,29 @@ * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. */ - #include +#include + +/* integer-arithmetic divisions of type (a * 1000000)/b */ +static inline u64 scaled_div(u64 a, u32 b) +{ + BUG_ON(b==0); + a *= 1000000; + do_div(a, b); + return a; +} + +static inline u32 scaled_div32(u64 a, u32 b) +{ + u64 result = scaled_div(a, b); + + if (result > UINT_MAX) { + DCCP_CRIT("Overflow: a(%llu)/b(%u) > ~0U", + (unsigned long long)a, b); + return UINT_MAX; + } + return result; +} extern u32 tfrc_calc_x(u16 s, u32 R, u32 p); extern u32 tfrc_calc_x_reverse_lookup(u32 fvalue); diff --git a/net/dccp/ccids/lib/tfrc_equation.c b/net/dccp/ccids/lib/tfrc_equation.c index ddac2c511e2f..90009fd77e15 100644 --- a/net/dccp/ccids/lib/tfrc_equation.c +++ b/net/dccp/ccids/lib/tfrc_equation.c @@ -13,7 +13,6 @@ */ #include -#include #include "../../dccp.h" #include "tfrc.h" @@ -616,15 +615,12 @@ static inline u32 tfrc_binsearch(u32 fval, u8 small) * @R: RTT scaled by 1000000 (i.e., microseconds) * @p: loss ratio estimate scaled by 1000000 * Returns X_calc in bytes per second (not scaled). - * - * Note: DO NOT alter this code unless you run test cases against it, - * as the code has been optimized to stop underflow/overflow. */ u32 tfrc_calc_x(u16 s, u32 R, u32 p) { - int index; + u16 index; u32 f; - u64 tmp1, tmp2; + u64 result; /* check against invalid parameters and divide-by-zero */ BUG_ON(p > 1000000); /* p must not exceed 100% */ @@ -650,15 +646,17 @@ u32 tfrc_calc_x(u16 s, u32 R, u32 p) f = tfrc_calc_x_lookup[index][0]; } - /* The following computes X = s/(R*f(p)) in bytes per second. Since f(p) - * and R are both scaled by 1000000, we need to multiply by 1000000^2. - * ==> DO NOT alter this unless you test against overflow on 32 bit */ - tmp1 = ((u64)s * 100000000); - tmp2 = ((u64)R * (u64)f); - do_div(tmp2, 10000); - do_div(tmp1, tmp2); - - return (u32)tmp1; + /* + * Compute X = s/(R*f(p)) in bytes per second. + * Since f(p) and R are both scaled by 1000000, we need to multiply by + * 1000000^2. To avoid overflow, the result is computed in two stages. + * This works under almost all reasonable operational conditions, for a + * wide range of parameters. Yet, should some strange combination of + * parameters result in overflow, the use of scaled_div32 will catch + * this and return UINT_MAX - which is a logically adequate consequence. + */ + result = scaled_div(s, R); + return scaled_div32(result, f); } EXPORT_SYMBOL_GPL(tfrc_calc_x); From bf58a381e8106fe73247c753e3da58fcb5eabd2e Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:04:43 -0200 Subject: [PATCH 15/32] [DCCP]: Only deliver to the CCID rx side in charge This is an optimisation to reduce CPU load. The received feedback is now only directed to the active CCID component, without requiring processing also by the inactive one. As a consequence, a similar test in ccid3.c is now redundant and is also removed. Justification: Currently DCCP works as a unidirectional service, i.e. a listening server is not at the same time a connecting client. As far as I can see, several modifications are necessary until that becomes possible. At the present time, received feedback is both fed to the rx/tx CCID modules. In unidirectional service, only one of these is active at any one time. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 6 +++--- net/dccp/input.c | 21 +++++++++++++++++---- 2 files changed, 20 insertions(+), 7 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 820bc25e14dc..936704d1b77e 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -540,11 +540,11 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) jiffies + usecs_to_jiffies(t_nfb)); /* set idle flag */ - hctx->ccid3hctx_idle = 1; + hctx->ccid3hctx_idle = 1; break; case TFRC_SSTATE_NO_SENT: - if (dccp_sk(sk)->dccps_role == DCCP_ROLE_CLIENT) - DCCP_WARN("Illegal ACK received - no packet sent\n"); + /* XXX when implementing bidirectional rx/tx check this again */ + DCCP_WARN("Illegal ACK received - no packet sent\n"); /* fall through */ case TFRC_SSTATE_TERM: /* ignore feedback when closing */ break; diff --git a/net/dccp/input.c b/net/dccp/input.c index 7371a2f3acf4..4a3279cd684c 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -248,8 +248,18 @@ int dccp_rcv_established(struct sock *sk, struct sk_buff *skb, DCCP_ACKVEC_STATE_RECEIVED)) goto discard; - ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); - ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb); + /* + * Deliver to the CCID module in charge. + * FIXME: Currently DCCP operates one-directional only, i.e. a listening + * server is not at the same time a connecting client. There is + * not much sense in delivering to both rx/tx sides at the moment + * (only one is active at a time); when moving to bidirectional + * service, this needs to be revised. + */ + if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER) + ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); + else + ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb); return __dccp_rcv_established(sk, skb, dh, len); discard: @@ -484,8 +494,11 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, DCCP_ACKVEC_STATE_RECEIVED)) goto discard; - ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); - ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb); + /* XXX see the comments in dccp_rcv_established about this */ + if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER) + ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); + else + ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb); } /* From f6282f4da5eb67e08eff94d73a1911c3a008846a Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:05:12 -0200 Subject: [PATCH 16/32] [DCCP]: Warn when discarding packet due to internal errors This adds a (debug) warning message which is triggered whenever a packet is discarded due to send failure. It also adds a conditional, so that an interruption during dccp_wait_for_ccid is not treated as a `BUG': the rationale is that interruptions are external, whereas bug warnings are concerned with the internals. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/output.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/dccp/output.c b/net/dccp/output.c index 29a8e8508462..b4df12b24161 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -241,7 +241,7 @@ void dccp_write_xmit(struct sock *sk, int block) break; } else err = dccp_wait_for_ccid(sk, skb); - if (err) + if (err && err != -EINTR) DCCP_BUG("err=%d after dccp_wait_for_ccid", err); } @@ -267,8 +267,10 @@ void dccp_write_xmit(struct sock *sk, int block) if (err) DCCP_BUG("err=%d after ccid_hc_tx_packet_sent", err); - } else + } else { + dccp_pr_debug("packet discarded\n"); kfree(skb); + } } } From 65d6c2b42ecc0ca77bb3c0fe27fcbdc84fc7a79a Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:05:38 -0200 Subject: [PATCH 17/32] [DCCP] ccid: Deprecate ccid_hc_tx_insert_options The function ccid3_hc_tx_insert_options only does a redundant no-op, as the operation DCCP_SKB_CB(skb)->dccpd_ccval = hctx->ccid3hctx_last_win_count; is already performed _unconditionally_ in ccid3_hc_tx_send_packet. Since there is further no current need for this function, it is removed entirely. Since furthermore, there is actually no present need for the entire interface function ccid_hc_tx_insert_options, it was decided to remove it also, to clean up the interface. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccid.h | 10 ---------- net/dccp/ccids/ccid3.c | 12 ------------ net/dccp/options.c | 5 ----- 3 files changed, 27 deletions(-) diff --git a/net/dccp/ccid.h b/net/dccp/ccid.h index bcc2d12ae81c..c65cb2453e43 100644 --- a/net/dccp/ccid.h +++ b/net/dccp/ccid.h @@ -43,8 +43,6 @@ struct ccid_operations { unsigned char* value); int (*ccid_hc_rx_insert_options)(struct sock *sk, struct sk_buff *skb); - int (*ccid_hc_tx_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, @@ -146,14 +144,6 @@ static inline int ccid_hc_rx_parse_options(struct ccid *ccid, struct sock *sk, return rc; } -static inline int ccid_hc_tx_insert_options(struct ccid *ccid, struct sock *sk, - struct sk_buff *skb) -{ - if (ccid->ccid_ops->ccid_hc_tx_insert_options != NULL) - return ccid->ccid_ops->ccid_hc_tx_insert_options(sk, skb); - return 0; -} - static inline int ccid_hc_rx_insert_options(struct ccid *ccid, struct sock *sk, struct sk_buff *skb) { diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 936704d1b77e..7017d84a381d 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -551,17 +551,6 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) } } -static int ccid3_hc_tx_insert_options(struct sock *sk, struct sk_buff *skb) -{ - const struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); - - BUG_ON(hctx == NULL); - - if (sk->sk_state == DCCP_OPEN || sk->sk_state == DCCP_PARTOPEN) - DCCP_SKB_CB(skb)->dccpd_ccval = hctx->ccid3hctx_last_win_count; - return 0; -} - static int ccid3_hc_tx_parse_options(struct sock *sk, unsigned char option, unsigned char len, u16 idx, unsigned char *value) @@ -1216,7 +1205,6 @@ static struct ccid_operations ccid3 = { .ccid_hc_tx_send_packet = ccid3_hc_tx_send_packet, .ccid_hc_tx_packet_sent = ccid3_hc_tx_packet_sent, .ccid_hc_tx_packet_recv = ccid3_hc_tx_packet_recv, - .ccid_hc_tx_insert_options = ccid3_hc_tx_insert_options, .ccid_hc_tx_parse_options = ccid3_hc_tx_parse_options, .ccid_hc_rx_obj_size = sizeof(struct ccid3_hc_rx_sock), .ccid_hc_rx_init = ccid3_hc_rx_init, diff --git a/net/dccp/options.c b/net/dccp/options.c index f398b43bc055..c03ba61eb6da 100644 --- a/net/dccp/options.c +++ b/net/dccp/options.c @@ -557,11 +557,6 @@ int dccp_insert_options(struct sock *sk, struct sk_buff *skb) return -1; dp->dccps_hc_rx_insert_options = 0; } - if (dp->dccps_hc_tx_insert_options) { - if (ccid_hc_tx_insert_options(dp->dccps_hc_tx_ccid, sk, skb)) - return -1; - dp->dccps_hc_tx_insert_options = 0; - } /* Feature negotiation */ /* Data packets can't do feat negotiation */ From fe0499ae95f5f636bda1f6e0bdba5b7b023ea827 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:06:01 -0200 Subject: [PATCH 18/32] [DCCP] ccid3: Initialise RTT values In both the sender and the receiver it is possible that the stored RTT value is accessed before an actual RTT estimate has been computed. This patch * initialises the sender RTT to 0 - the sender always accesses the RTT in ccid3_hc_tx_packet_sent - the RTT is further needed for the window counter algorithm * replaces the receiver initialisation of 5msec with 0 - which has the same effect and removes an `XXX' - the RTT value is needed in ccid3_hc_rx_packet_recv as rtt_prev Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 7017d84a381d..7618d51aa2d2 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -617,6 +617,7 @@ static int ccid3_hc_tx_init(struct ccid *ccid, struct sock *sk) struct ccid3_hc_tx_sock *hctx = ccid_priv(ccid); hctx->ccid3hctx_s = 0; + hctx->ccid3hctx_rtt = 0; hctx->ccid3hctx_state = TFRC_SSTATE_NO_SENT; INIT_LIST_HEAD(&hctx->ccid3hctx_hist); @@ -1093,7 +1094,7 @@ static int ccid3_hc_rx_init(struct ccid *ccid, struct sock *sk) dccp_timestamp(sk, &hcrx->ccid3hcrx_tstamp_last_ack); hcrx->ccid3hcrx_tstamp_last_feedback = hcrx->ccid3hcrx_tstamp_last_ack; hcrx->ccid3hcrx_s = 0; - hcrx->ccid3hcrx_rtt = 5000; /* XXX 5ms for now... */ + hcrx->ccid3hcrx_rtt = 0; return 0; } From de553c189e3faa0d0c38f366f73379b46587b80e Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:06:32 -0200 Subject: [PATCH 19/32] [DCCP] ccid3: Sanity-check RTT samples CCID3 performance depends much on the accuracy of RTT samples. If RTT samples grow too large, performance can be catastrophically poor. To limit the amount of possible damage in such cases, the patch * introduces an upper limit which identifies a maximum `sane' RTT value; * uses a macro to enforce this upper limit. Using a macro was given preference, since it is necessary to identify the calling function in the warning message. Since exceeding this threshold identifies a critical condition, DCCP_CRIT is used and not DCCP_WARN. Many thanks to Ian McDonald for collaboration on this issue. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 4 +++- net/dccp/ccids/ccid3.h | 10 ++++++++++ 2 files changed, 13 insertions(+), 1 deletion(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 7618d51aa2d2..122a716eb877 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -456,8 +456,9 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) r_sample, t_elapsed); else r_sample -= t_elapsed; + CCID3_RTT_SANITY_CHECK(r_sample); - /* Update RTT estimate by + /* Update RTT estimate by * If (No feedback recv) * R = R_sample; * Else @@ -1000,6 +1001,7 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) r_sample, t_elapsed); else r_sample -= t_elapsed; + CCID3_RTT_SANITY_CHECK(r_sample); if (hcrx->ccid3hcrx_state == TFRC_RSTATE_NO_DATA) hcrx->ccid3hcrx_rtt = r_sample; diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index da0ca3c0a7b3..3fa0f69ed320 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -51,6 +51,16 @@ /* Parameter t_mbi from [RFC 3448, 4.3]: backoff interval in seconds */ #define TFRC_T_MBI 64 +/* What we think is a reasonable upper limit on RTT values */ +#define CCID3_SANE_RTT_MAX (4 * USEC_PER_SEC) + +#define CCID3_RTT_SANITY_CHECK(rtt) do { \ + if (rtt > CCID3_SANE_RTT_MAX) { \ + DCCP_CRIT("RTT (%ld) too large, substituting %ld", \ + rtt, CCID3_SANE_RTT_MAX); \ + rtt = CCID3_SANE_RTT_MAX; \ + } } while (0) + enum ccid3_options { TFRC_OPT_LOSS_EVENT_RATE = 192, TFRC_OPT_LOSS_INTERVALS = 193, From 9f8681db961de6d75e43844e9664f1f81e05e1b7 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:07:37 -0200 Subject: [PATCH 20/32] [DCCP] ccid3: Shift window counter computation This puts the window counter computation [RFC 4342, 8.1] into a separate function which is called whenever a new packet is ready for immediate transmission in ccid3_hc_tx_send_packet. Justification: The window counter update was previously computed after the packet was sent. This has two drawbacks, both fixed by this patch: 1) re-compute another timestamp almost directly after the packet was sent (expensive), 2) the CCVal for the window counter is needed at the instant the packet is sent. Further details: The initialisation of the window counter is left in the state NO_SENT, as before. The algorithm will do nothing if either RTT is initialised to 0 (which is ok) or if the RTT value remains below 4 microseconds (which is almost pathological). Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 49 +++++++++++++++++++++++++----------------- 1 file changed, 29 insertions(+), 20 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 122a716eb877..1a2ffa1d8794 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -161,6 +161,33 @@ static inline void ccid3_hc_tx_update_s(struct ccid3_hc_tx_sock *hctx, int len) */ } +/* + * Update Window Counter using the algorithm from [RFC 4342, 8.1]. + * The algorithm is not applicable if RTT < 4 microseconds. + */ +static inline void ccid3_hc_tx_update_win_count(struct ccid3_hc_tx_sock *hctx, + struct timeval *now) +{ + suseconds_t delta; + u32 quarter_rtts; + + if (unlikely(hctx->ccid3hctx_rtt < 4)) /* avoid divide-by-zero */ + return; + + delta = timeval_delta(now, &hctx->ccid3hctx_t_last_win_count); + DCCP_BUG_ON(delta < 0); + + quarter_rtts = (u32)delta / (hctx->ccid3hctx_rtt / 4); + + if (quarter_rtts > 0) { + hctx->ccid3hctx_t_last_win_count = *now; + hctx->ccid3hctx_last_win_count += min_t(u32, quarter_rtts, 5); + hctx->ccid3hctx_last_win_count &= 0xF; /* mod 16 */ + + ccid3_pr_debug("now at %#X\n", hctx->ccid3hctx_last_win_count); + } +} + static void ccid3_hc_tx_no_feedback_timer(unsigned long data) { struct sock *sk = (struct sock *)data; @@ -333,6 +360,8 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) */ if (delay - (suseconds_t)hctx->ccid3hctx_delta >= 0) return delay / 1000L; + + ccid3_hc_tx_update_win_count(hctx, &now); break; case TFRC_SSTATE_TERM: DCCP_BUG("Illegal %s state TERM, sk=%p", dccp_role(sk), sk); @@ -353,7 +382,6 @@ static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) const struct dccp_sock *dp = dccp_sk(sk); struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); struct timeval now; - suseconds_t quarter_rtt; struct dccp_tx_hist_entry *packet; BUG_ON(hctx == NULL); @@ -373,25 +401,6 @@ static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) } packet->dccphtx_tstamp = now; packet->dccphtx_seqno = dp->dccps_gss; - /* - * Check if win_count have changed - * Algorithm in "8.1. Window Counter Value" in RFC 4342. - */ - quarter_rtt = timeval_delta(&now, &hctx->ccid3hctx_t_last_win_count); - if (likely(hctx->ccid3hctx_rtt > 8)) - quarter_rtt /= hctx->ccid3hctx_rtt / 4; - - if (quarter_rtt > 0) { - hctx->ccid3hctx_t_last_win_count = now; - hctx->ccid3hctx_last_win_count = (hctx->ccid3hctx_last_win_count + - min_t(unsigned long, quarter_rtt, 5)) % 16; - ccid3_pr_debug("%s, sk=%p, window changed from " - "%u to %u!\n", - dccp_role(sk), sk, - packet->dccphtx_ccval, - hctx->ccid3hctx_last_win_count); - } - hctx->ccid3hctx_idle = 0; packet->dccphtx_rtt = hctx->ccid3hctx_rtt; packet->dccphtx_sent = 1; From e312d100f19fdfe1019512b07a9d15653f254abf Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:08:09 -0200 Subject: [PATCH 21/32] [DCCP] ccid3: TX history - remove unused field This removes the `dccphtx_ccval' field since it is nowhere used in the code and in fact not necessary for the accounting. Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 5 +++-- net/dccp/ccids/lib/packet_history.h | 1 - 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 1a2ffa1d8794..6bb3f8352dad 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -370,8 +370,9 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) /* prepare to send now (add options etc.) */ dp->dccps_hc_tx_insert_options = 1; - new_packet->dccphtx_ccval = DCCP_SKB_CB(skb)->dccpd_ccval = - hctx->ccid3hctx_last_win_count; + DCCP_SKB_CB(skb)->dccpd_ccval = hctx->ccid3hctx_last_win_count; + + /* set the nominal send time for the next following packet */ timeval_add_usecs(&hctx->ccid3hctx_t_nom, hctx->ccid3hctx_t_ipi); return 0; diff --git a/net/dccp/ccids/lib/packet_history.h b/net/dccp/ccids/lib/packet_history.h index 9a8bcf224aa7..4198185c6d40 100644 --- a/net/dccp/ccids/lib/packet_history.h +++ b/net/dccp/ccids/lib/packet_history.h @@ -52,7 +52,6 @@ struct dccp_tx_hist_entry { struct list_head dccphtx_node; u64 dccphtx_seqno:48, - dccphtx_ccval:4, dccphtx_sent:1; u32 dccphtx_rtt; struct timeval dccphtx_tstamp; From c5a1ae9a4cc4aef1505fa0aba079f834c3ee1af1 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:09:21 -0200 Subject: [PATCH 22/32] [DCCP] ccid3: Perform history operations only after packet has been sent This migrates all packet history operations into the routine ccid3_hc_tx_packet_sent, thereby removing synchronization problems that occur when, as before, the operations are spread over multiple routines. The following minor simplifications are also applied: * several simplifications now follow from this change - several tests are now no longer required * removal of one unnecessary variable (dp) Justification: Currently packet history operations span two different routines, one of which is likely to pass through several iterations of sleeping and awakening. The first routine, ccid3_hc_tx_send_packet, allocates an entry and sets a few fields. The remaining fields are filled in when the second routine (which is not within a sleeping context), ccid3_hc_tx_packet_sent, is called. This has several strong drawbacks: * it is not necessary to split history operations - all fields can be filled in by the second routine * the first routine is called multiple times, until a packet can be sent, and sleeps meanwhile - this causes a lot of difficulties with regard to keeping the list consistent * since both routines do not have a producer-consumer like synchronization, it is very difficult to maintain data across calls to these routines * the fact that the routines are called in different contexts (sleeping, not sleeping) adds further problems Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 38 +++++++++----------------------------- 1 file changed, 9 insertions(+), 29 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 6bb3f8352dad..ef68c35183ed 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -296,7 +296,6 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) { struct dccp_sock *dp = dccp_sk(sk); struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); - struct dccp_tx_hist_entry *new_packet; struct timeval now; suseconds_t delay; @@ -310,21 +309,6 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) if (unlikely(skb->len == 0)) return -EBADMSG; - /* See if last packet allocated was not sent */ - new_packet = dccp_tx_hist_head(&hctx->ccid3hctx_hist); - if (new_packet == NULL || new_packet->dccphtx_sent) { - new_packet = dccp_tx_hist_entry_new(ccid3_tx_hist, - GFP_ATOMIC); - - if (unlikely(new_packet == NULL)) { - DCCP_WARN("%s, sk=%p, not enough mem to add to history," - "send refused\n", dccp_role(sk), sk); - return -ENOBUFS; - } - - dccp_tx_hist_add_entry(&hctx->ccid3hctx_hist, new_packet); - } - dccp_timestamp(sk, &now); switch (hctx->ccid3hctx_state) { @@ -380,31 +364,27 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) { - const struct dccp_sock *dp = dccp_sk(sk); struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); struct timeval now; struct dccp_tx_hist_entry *packet; BUG_ON(hctx == NULL); - dccp_timestamp(sk, &now); - ccid3_hc_tx_update_s(hctx, len); - packet = dccp_tx_hist_head(&hctx->ccid3hctx_hist); + packet = dccp_tx_hist_entry_new(ccid3_tx_hist, GFP_ATOMIC); if (unlikely(packet == NULL)) { - DCCP_WARN("packet doesn't exist in history!\n"); - return; - } - if (unlikely(packet->dccphtx_sent)) { - DCCP_WARN("no unsent packet in history!\n"); + DCCP_CRIT("packet history - out of memory!"); return; } + dccp_tx_hist_add_entry(&hctx->ccid3hctx_hist, packet); + + dccp_timestamp(sk, &now); packet->dccphtx_tstamp = now; - packet->dccphtx_seqno = dp->dccps_gss; - hctx->ccid3hctx_idle = 0; - packet->dccphtx_rtt = hctx->ccid3hctx_rtt; - packet->dccphtx_sent = 1; + packet->dccphtx_seqno = dccp_sk(sk)->dccps_gss; + packet->dccphtx_rtt = hctx->ccid3hctx_rtt; + packet->dccphtx_sent = 1; + hctx->ccid3hctx_idle = 0; } static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) From a967241129f6dc1db92fee2c808f73a90f5f47a3 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:14:12 -0200 Subject: [PATCH 23/32] [DCCP] ccid3: Make debug output consistent This patch does not alter any algorithm, just the debug message format: * s#%s, sk=%p#%s(%p)#g * when a statename is present, it now uses %s(%p, state=%s) * when only function entry is debugged, it adds an `- entry' Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 67 ++++++++++++++++++++---------------------- 1 file changed, 32 insertions(+), 35 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index ef68c35183ed..d23af7b475a0 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -201,18 +201,17 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) goto restart_timer; } - ccid3_pr_debug("%s, sk=%p, state=%s\n", dccp_role(sk), sk, + ccid3_pr_debug("%s(%p, state=%s) - entry \n", dccp_role(sk), sk, ccid3_tx_state_name(hctx->ccid3hctx_state)); - + switch (hctx->ccid3hctx_state) { case TFRC_SSTATE_NO_FBACK: /* RFC 3448, 4.4: Halve send rate directly */ hctx->ccid3hctx_x = max_t(u32, hctx->ccid3hctx_x / 2, (hctx->ccid3hctx_s << 6)/TFRC_T_MBI); - ccid3_pr_debug("%s, sk=%p, state=%s, updated tx rate to %u " - "bytes/s\n", - dccp_role(sk), sk, + ccid3_pr_debug("%s(%p, state=%s), updated tx rate to %u " + "bytes/s\n", dccp_role(sk), sk, ccid3_tx_state_name(hctx->ccid3hctx_state), (unsigned)(hctx->ccid3hctx_x >> 6)); /* The value of R is still undefined and so we can not recompute @@ -230,7 +229,7 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) scaled_div(hctx->ccid3hctx_s << 6, hctx->ccid3hctx_rtt))) { struct timeval now; - ccid3_pr_debug("%s, sk=%p, state=%s, not idle\n", + ccid3_pr_debug("%s(%p, state=%s), not idle\n", dccp_role(sk), sk, ccid3_tx_state_name(hctx->ccid3hctx_state)); @@ -270,7 +269,7 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) t_nfb = max(hctx->ccid3hctx_t_rto, 2 * hctx->ccid3hctx_t_ipi); break; case TFRC_SSTATE_NO_SENT: - DCCP_BUG("Illegal %s state NO_SENT, sk=%p", dccp_role(sk), sk); + DCCP_BUG("%s(%p) - Illegal state NO_SENT", dccp_role(sk), sk); /* fall through */ case TFRC_SSTATE_TERM: goto out; @@ -348,7 +347,7 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) ccid3_hc_tx_update_win_count(hctx, &now); break; case TFRC_SSTATE_TERM: - DCCP_BUG("Illegal %s state TERM, sk=%p", dccp_role(sk), sk); + DCCP_BUG("%s(%p) - Illegal state TERM", dccp_role(sk), sk); return -EINVAL; } @@ -521,13 +520,12 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) * max(t_RTO, 2 * s/X) = max(t_RTO, 2 * t_ipi) */ t_nfb = max(hctx->ccid3hctx_t_rto, 2 * hctx->ccid3hctx_t_ipi); - - ccid3_pr_debug("%s, sk=%p, Scheduled no feedback timer to " - "expire in %lu jiffies (%luus)\n", - dccp_role(sk), sk, - usecs_to_jiffies(t_nfb), t_nfb); - sk_reset_timer(sk, &hctx->ccid3hctx_no_feedback_timer, + ccid3_pr_debug("%s(%p), Scheduled no feedback timer to " + "expire in %lu jiffies (%luus)\n", dccp_role(sk), + sk, usecs_to_jiffies(t_nfb), t_nfb); + + sk_reset_timer(sk, &hctx->ccid3hctx_no_feedback_timer, jiffies + usecs_to_jiffies(t_nfb)); /* set idle flag */ @@ -566,13 +564,13 @@ static int ccid3_hc_tx_parse_options(struct sock *sk, unsigned char option, switch (option) { case TFRC_OPT_LOSS_EVENT_RATE: if (unlikely(len != 4)) { - DCCP_WARN("%s, sk=%p, invalid len %d " + DCCP_WARN("%s(%p), invalid len %d " "for TFRC_OPT_LOSS_EVENT_RATE\n", dccp_role(sk), sk, len); rc = -EINVAL; } else { opt_recv->ccid3or_loss_event_rate = ntohl(*(__be32 *)value); - ccid3_pr_debug("%s, sk=%p, LOSS_EVENT_RATE=%u\n", + ccid3_pr_debug("%s(%p), LOSS_EVENT_RATE=%u\n", dccp_role(sk), sk, opt_recv->ccid3or_loss_event_rate); } @@ -580,20 +578,20 @@ static int ccid3_hc_tx_parse_options(struct sock *sk, unsigned char option, case TFRC_OPT_LOSS_INTERVALS: opt_recv->ccid3or_loss_intervals_idx = idx; opt_recv->ccid3or_loss_intervals_len = len; - ccid3_pr_debug("%s, sk=%p, LOSS_INTERVALS=(%u, %u)\n", + ccid3_pr_debug("%s(%p), LOSS_INTERVALS=(%u, %u)\n", dccp_role(sk), sk, opt_recv->ccid3or_loss_intervals_idx, opt_recv->ccid3or_loss_intervals_len); break; case TFRC_OPT_RECEIVE_RATE: if (unlikely(len != 4)) { - DCCP_WARN("%s, sk=%p, invalid len %d " + DCCP_WARN("%s(%p), invalid len %d " "for TFRC_OPT_RECEIVE_RATE\n", dccp_role(sk), sk, len); rc = -EINVAL; } else { opt_recv->ccid3or_receive_rate = ntohl(*(__be32 *)value); - ccid3_pr_debug("%s, sk=%p, RECEIVE_RATE=%u\n", + ccid3_pr_debug("%s(%p), RECEIVE_RATE=%u\n", dccp_role(sk), sk, opt_recv->ccid3or_receive_rate); } @@ -679,7 +677,7 @@ static void ccid3_hc_rx_send_feedback(struct sock *sk) struct timeval now; suseconds_t delta; - ccid3_pr_debug("%s, sk=%p\n", dccp_role(sk), sk); + ccid3_pr_debug("%s(%p) - entry \n", dccp_role(sk), sk); dccp_timestamp(sk, &now); @@ -695,13 +693,13 @@ static void ccid3_hc_rx_send_feedback(struct sock *sk) scaled_div32(hcrx->ccid3hcrx_bytes_recv, delta); break; case TFRC_RSTATE_TERM: - DCCP_BUG("Illegal %s state TERM, sk=%p", dccp_role(sk), sk); + DCCP_BUG("%s(%p) - Illegal state TERM", dccp_role(sk), sk); return; } packet = dccp_rx_hist_find_data_packet(&hcrx->ccid3hcrx_hist); if (unlikely(packet == NULL)) { - DCCP_WARN("%s, sk=%p, no data packet in history!\n", + DCCP_WARN("%s(%p), no data packet in history!\n", dccp_role(sk), sk); return; } @@ -797,13 +795,13 @@ static u32 ccid3_hc_rx_calc_first_li(struct sock *sk) } if (unlikely(step == 0)) { - DCCP_WARN("%s, sk=%p, packet history has no data packets!\n", + DCCP_WARN("%s(%p), packet history has no data packets!\n", dccp_role(sk), sk); return ~0; } if (unlikely(interval == 0)) { - DCCP_WARN("%s, sk=%p, Could not find a win_count interval > 0." + DCCP_WARN("%s(%p), Could not find a win_count interval > 0." "Defaulting to 1\n", dccp_role(sk), sk); interval = 1; } @@ -817,7 +815,7 @@ found: DCCP_BUG_ON(delta < 0); rtt = delta * 4 / interval; - ccid3_pr_debug("%s, sk=%p, approximated RTT to %ldus\n", + ccid3_pr_debug("%s(%p), approximated RTT to %ldus\n", dccp_role(sk), sk, rtt); /* @@ -850,7 +848,7 @@ found: fval = scaled_div32(fval, x_recv); p = tfrc_calc_x_reverse_lookup(fval); - ccid3_pr_debug("%s, sk=%p, receive rate=%u bytes/s, implied " + ccid3_pr_debug("%s(%p), receive rate=%u bytes/s, implied " "loss rate=%u\n", dccp_role(sk), sk, x_recv, p); if (p == 0) @@ -1000,8 +998,8 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) r_sample / 10; if (rtt_prev != hcrx->ccid3hcrx_rtt) - ccid3_pr_debug("%s, New RTT=%uus, elapsed time=%u\n", - dccp_role(sk), hcrx->ccid3hcrx_rtt, + ccid3_pr_debug("%s(%p), New RTT=%uus, elapsed time=%u\n", + dccp_role(sk), sk, hcrx->ccid3hcrx_rtt, opt_recv->dccpor_elapsed_time); break; case DCCP_PKT_DATA: @@ -1013,7 +1011,7 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) packet = dccp_rx_hist_entry_new(ccid3_rx_hist, sk, opt_recv->dccpor_ndp, skb, GFP_ATOMIC); if (unlikely(packet == NULL)) { - DCCP_WARN("%s, sk=%p, Not enough mem to add rx packet " + DCCP_WARN("%s(%p), Not enough mem to add rx packet " "to history, consider it lost!\n", dccp_role(sk), sk); return; } @@ -1028,9 +1026,8 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) switch (hcrx->ccid3hcrx_state) { case TFRC_RSTATE_NO_DATA: - ccid3_pr_debug("%s, sk=%p(%s), skb=%p, sending initial " - "feedback\n", - dccp_role(sk), sk, + ccid3_pr_debug("%s(%p, state=%s), skb=%p, sending initial " + "feedback\n", dccp_role(sk), sk, dccp_state_name(sk->sk_state), skb); ccid3_hc_rx_send_feedback(sk); ccid3_hc_rx_set_state(sk, TFRC_RSTATE_DATA); @@ -1048,12 +1045,12 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) } return; case TFRC_RSTATE_TERM: - DCCP_BUG("Illegal %s state TERM, sk=%p", dccp_role(sk), sk); + DCCP_BUG("%s(%p) - Illegal state TERM", dccp_role(sk), sk); return; } /* Dealing with packet loss */ - ccid3_pr_debug("%s, sk=%p(%s), data loss! Reacting...\n", + ccid3_pr_debug("%s(%p, state=%s), data loss! Reacting...\n", dccp_role(sk), sk, dccp_state_name(sk->sk_state)); p_prev = hcrx->ccid3hcrx_p; @@ -1078,7 +1075,7 @@ static int ccid3_hc_rx_init(struct ccid *ccid, struct sock *sk) { struct ccid3_hc_rx_sock *hcrx = ccid_priv(ccid); - ccid3_pr_debug("%s, sk=%p\n", dccp_role(sk), sk); + ccid3_pr_debug("entry\n"); hcrx->ccid3hcrx_state = TFRC_RSTATE_NO_DATA; INIT_LIST_HEAD(&hcrx->ccid3hcrx_hist); From 85dcb1f78039cc55ccddd9622657a338b718aafe Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:24:11 -0200 Subject: [PATCH 24/32] [DCCP] ccid3: Reorder packet history header file No code change at all. To make the header file easier to read, the following ordering is established among the declarations: * hist_new * hist_delete * hist_entry_new * hist_head * hist_find_entry * hist_add_entry * hist_entry_delete * hist_purge Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/lib/packet_history.h | 147 +++++++++++++++------------- 1 file changed, 77 insertions(+), 70 deletions(-) diff --git a/net/dccp/ccids/lib/packet_history.h b/net/dccp/ccids/lib/packet_history.h index 4198185c6d40..1f960c19ea1b 100644 --- a/net/dccp/ccids/lib/packet_history.h +++ b/net/dccp/ccids/lib/packet_history.h @@ -49,6 +49,9 @@ #define TFRC_WIN_COUNT_PER_RTT 4 #define TFRC_WIN_COUNT_LIMIT 16 +/* + * Transmitter History data structures and declarations + */ struct dccp_tx_hist_entry { struct list_head dccphtx_node; u64 dccphtx_seqno:48, @@ -57,34 +60,16 @@ struct dccp_tx_hist_entry { struct timeval dccphtx_tstamp; }; -struct dccp_rx_hist_entry { - struct list_head dccphrx_node; - u64 dccphrx_seqno:48, - dccphrx_ccval:4, - dccphrx_type:4; - u32 dccphrx_ndp; /* In fact it is from 8 to 24 bits */ - struct timeval dccphrx_tstamp; -}; - struct dccp_tx_hist { struct kmem_cache *dccptxh_slab; }; extern struct dccp_tx_hist *dccp_tx_hist_new(const char *name); -extern void dccp_tx_hist_delete(struct dccp_tx_hist *hist); - -struct dccp_rx_hist { - struct kmem_cache *dccprxh_slab; -}; - -extern struct dccp_rx_hist *dccp_rx_hist_new(const char *name); -extern void dccp_rx_hist_delete(struct dccp_rx_hist *hist); -extern struct dccp_rx_hist_entry * - dccp_rx_hist_find_data_packet(const struct list_head *list); +extern void dccp_tx_hist_delete(struct dccp_tx_hist *hist); static inline struct dccp_tx_hist_entry * - dccp_tx_hist_entry_new(struct dccp_tx_hist *hist, - const gfp_t prio) + dccp_tx_hist_entry_new(struct dccp_tx_hist *hist, + const gfp_t prio) { struct dccp_tx_hist_entry *entry = kmem_cache_alloc(hist->dccptxh_slab, prio); @@ -95,34 +80,8 @@ static inline struct dccp_tx_hist_entry * return entry; } -static inline void dccp_tx_hist_entry_delete(struct dccp_tx_hist *hist, - struct dccp_tx_hist_entry *entry) -{ - if (entry != NULL) - kmem_cache_free(hist->dccptxh_slab, entry); -} - -extern struct dccp_tx_hist_entry * - dccp_tx_hist_find_entry(const struct list_head *list, - const u64 seq); -extern int dccp_rx_hist_find_entry(const struct list_head *list, const u64 seq, - u8 *ccval); - -static inline void dccp_tx_hist_add_entry(struct list_head *list, - struct dccp_tx_hist_entry *entry) -{ - list_add(&entry->dccphtx_node, list); -} - -extern void dccp_tx_hist_purge_older(struct dccp_tx_hist *hist, - struct list_head *list, - struct dccp_tx_hist_entry *next); - -extern void dccp_tx_hist_purge(struct dccp_tx_hist *hist, - struct list_head *list); - static inline struct dccp_tx_hist_entry * - dccp_tx_hist_head(struct list_head *list) + dccp_tx_hist_head(struct list_head *list) { struct dccp_tx_hist_entry *head = NULL; @@ -132,12 +91,55 @@ static inline struct dccp_tx_hist_entry * return head; } +extern struct dccp_tx_hist_entry * + dccp_tx_hist_find_entry(const struct list_head *list, + const u64 seq); + +static inline void dccp_tx_hist_add_entry(struct list_head *list, + struct dccp_tx_hist_entry *entry) +{ + list_add(&entry->dccphtx_node, list); +} + +static inline void dccp_tx_hist_entry_delete(struct dccp_tx_hist *hist, + struct dccp_tx_hist_entry *entry) +{ + if (entry != NULL) + kmem_cache_free(hist->dccptxh_slab, entry); +} + +extern void dccp_tx_hist_purge(struct dccp_tx_hist *hist, + struct list_head *list); + +extern void dccp_tx_hist_purge_older(struct dccp_tx_hist *hist, + struct list_head *list, + struct dccp_tx_hist_entry *next); + +/* + * Receiver History data structures and declarations + */ +struct dccp_rx_hist_entry { + struct list_head dccphrx_node; + u64 dccphrx_seqno:48, + dccphrx_ccval:4, + dccphrx_type:4; + u32 dccphrx_ndp; /* In fact it is from 8 to 24 bits */ + struct timeval dccphrx_tstamp; +}; + +struct dccp_rx_hist { + struct kmem_cache *dccprxh_slab; +}; + +extern struct dccp_rx_hist *dccp_rx_hist_new(const char *name); +extern void dccp_rx_hist_delete(struct dccp_rx_hist *hist); + static inline struct dccp_rx_hist_entry * - dccp_rx_hist_entry_new(struct dccp_rx_hist *hist, - const struct sock *sk, - const u32 ndp, - const struct sk_buff *skb, - const gfp_t prio) + dccp_rx_hist_entry_new(struct dccp_rx_hist *hist, + const struct sock *sk, + const u32 ndp, + const struct sk_buff *skb, + const gfp_t prio) { struct dccp_rx_hist_entry *entry = kmem_cache_alloc(hist->dccprxh_slab, prio); @@ -155,6 +157,28 @@ static inline struct dccp_rx_hist_entry * return entry; } +static inline struct dccp_rx_hist_entry * + dccp_rx_hist_head(struct list_head *list) +{ + struct dccp_rx_hist_entry *head = NULL; + + if (!list_empty(list)) + head = list_entry(list->next, struct dccp_rx_hist_entry, + dccphrx_node); + return head; +} + +extern int dccp_rx_hist_find_entry(const struct list_head *list, const u64 seq, + u8 *ccval); +extern struct dccp_rx_hist_entry * + dccp_rx_hist_find_data_packet(const struct list_head *list); + +extern void dccp_rx_hist_add_packet(struct dccp_rx_hist *hist, + struct list_head *rx_list, + struct list_head *li_list, + struct dccp_rx_hist_entry *packet, + u64 nonloss_seqno); + static inline void dccp_rx_hist_entry_delete(struct dccp_rx_hist *hist, struct dccp_rx_hist_entry *entry) { @@ -165,17 +189,6 @@ static inline void dccp_rx_hist_entry_delete(struct dccp_rx_hist *hist, extern void dccp_rx_hist_purge(struct dccp_rx_hist *hist, struct list_head *list); -static inline struct dccp_rx_hist_entry * - dccp_rx_hist_head(struct list_head *list) -{ - struct dccp_rx_hist_entry *head = NULL; - - if (!list_empty(list)) - head = list_entry(list->next, struct dccp_rx_hist_entry, - dccphrx_node); - return head; -} - static inline int dccp_rx_hist_entry_data_packet(const struct dccp_rx_hist_entry *entry) { @@ -183,12 +196,6 @@ static inline int entry->dccphrx_type == DCCP_PKT_DATAACK; } -extern void dccp_rx_hist_add_packet(struct dccp_rx_hist *hist, - struct list_head *rx_list, - struct list_head *li_list, - struct dccp_rx_hist_entry *packet, - u64 nonloss_seqno); - extern u64 dccp_rx_hist_detect_loss(struct list_head *rx_list, struct list_head *li_list, u8 *win_loss); From 7af5af3013f84693a7af581e3c8b32f9db493cf5 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:24:33 -0200 Subject: [PATCH 25/32] [DCCP] ccid3: Reorder packet history source file No code change at all. This reorders the source file to follow the same order as the corresponding header file. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/lib/packet_history.c | 219 ++++++++++++++-------------- 1 file changed, 112 insertions(+), 107 deletions(-) diff --git a/net/dccp/ccids/lib/packet_history.c b/net/dccp/ccids/lib/packet_history.c index b876c9c81c65..2e8ef42721e2 100644 --- a/net/dccp/ccids/lib/packet_history.c +++ b/net/dccp/ccids/lib/packet_history.c @@ -36,9 +36,100 @@ #include #include - #include "packet_history.h" +/* + * Transmitter History Routines + */ +struct dccp_tx_hist *dccp_tx_hist_new(const char *name) +{ + struct dccp_tx_hist *hist = kmalloc(sizeof(*hist), GFP_ATOMIC); + static const char dccp_tx_hist_mask[] = "tx_hist_%s"; + char *slab_name; + + if (hist == NULL) + goto out; + + slab_name = kmalloc(strlen(name) + sizeof(dccp_tx_hist_mask) - 1, + GFP_ATOMIC); + if (slab_name == NULL) + goto out_free_hist; + + sprintf(slab_name, dccp_tx_hist_mask, name); + hist->dccptxh_slab = kmem_cache_create(slab_name, + sizeof(struct dccp_tx_hist_entry), + 0, SLAB_HWCACHE_ALIGN, + NULL, NULL); + if (hist->dccptxh_slab == NULL) + goto out_free_slab_name; +out: + return hist; +out_free_slab_name: + kfree(slab_name); +out_free_hist: + kfree(hist); + hist = NULL; + goto out; +} + +EXPORT_SYMBOL_GPL(dccp_tx_hist_new); + +void dccp_tx_hist_delete(struct dccp_tx_hist *hist) +{ + const char* name = kmem_cache_name(hist->dccptxh_slab); + + kmem_cache_destroy(hist->dccptxh_slab); + kfree(name); + kfree(hist); +} + +EXPORT_SYMBOL_GPL(dccp_tx_hist_delete); + +struct dccp_tx_hist_entry * + dccp_tx_hist_find_entry(const struct list_head *list, const u64 seq) +{ + struct dccp_tx_hist_entry *packet = NULL, *entry; + + list_for_each_entry(entry, list, dccphtx_node) + if (entry->dccphtx_seqno == seq) { + packet = entry; + break; + } + + return packet; +} + +EXPORT_SYMBOL_GPL(dccp_tx_hist_find_entry); + +void dccp_tx_hist_purge(struct dccp_tx_hist *hist, struct list_head *list) +{ + struct dccp_tx_hist_entry *entry, *next; + + list_for_each_entry_safe(entry, next, list, dccphtx_node) { + list_del_init(&entry->dccphtx_node); + dccp_tx_hist_entry_delete(hist, entry); + } +} + +EXPORT_SYMBOL_GPL(dccp_tx_hist_purge); + +void dccp_tx_hist_purge_older(struct dccp_tx_hist *hist, + struct list_head *list, + struct dccp_tx_hist_entry *packet) +{ + struct dccp_tx_hist_entry *next; + + list_for_each_entry_safe_continue(packet, next, list, dccphtx_node) { + list_del_init(&packet->dccphtx_node); + dccp_tx_hist_entry_delete(hist, packet); + } +} + +EXPORT_SYMBOL_GPL(dccp_tx_hist_purge_older); + +/* + * Receiver History Routines + */ struct dccp_rx_hist *dccp_rx_hist_new(const char *name) { struct dccp_rx_hist *hist = kmalloc(sizeof(*hist), GFP_ATOMIC); @@ -83,18 +174,24 @@ void dccp_rx_hist_delete(struct dccp_rx_hist *hist) EXPORT_SYMBOL_GPL(dccp_rx_hist_delete); -void dccp_rx_hist_purge(struct dccp_rx_hist *hist, struct list_head *list) +int dccp_rx_hist_find_entry(const struct list_head *list, const u64 seq, + u8 *ccval) { - struct dccp_rx_hist_entry *entry, *next; + struct dccp_rx_hist_entry *packet = NULL, *entry; - list_for_each_entry_safe(entry, next, list, dccphrx_node) { - list_del_init(&entry->dccphrx_node); - kmem_cache_free(hist->dccprxh_slab, entry); - } + list_for_each_entry(entry, list, dccphrx_node) + if (entry->dccphrx_seqno == seq) { + packet = entry; + break; + } + + if (packet) + *ccval = packet->dccphrx_ccval; + + return packet != NULL; } -EXPORT_SYMBOL_GPL(dccp_rx_hist_purge); - +EXPORT_SYMBOL_GPL(dccp_rx_hist_find_entry); struct dccp_rx_hist_entry * dccp_rx_hist_find_data_packet(const struct list_head *list) { @@ -184,110 +281,18 @@ void dccp_rx_hist_add_packet(struct dccp_rx_hist *hist, EXPORT_SYMBOL_GPL(dccp_rx_hist_add_packet); -struct dccp_tx_hist *dccp_tx_hist_new(const char *name) +void dccp_rx_hist_purge(struct dccp_rx_hist *hist, struct list_head *list) { - struct dccp_tx_hist *hist = kmalloc(sizeof(*hist), GFP_ATOMIC); - static const char dccp_tx_hist_mask[] = "tx_hist_%s"; - char *slab_name; + struct dccp_rx_hist_entry *entry, *next; - if (hist == NULL) - goto out; - - slab_name = kmalloc(strlen(name) + sizeof(dccp_tx_hist_mask) - 1, - GFP_ATOMIC); - if (slab_name == NULL) - goto out_free_hist; - - sprintf(slab_name, dccp_tx_hist_mask, name); - hist->dccptxh_slab = kmem_cache_create(slab_name, - sizeof(struct dccp_tx_hist_entry), - 0, SLAB_HWCACHE_ALIGN, - NULL, NULL); - if (hist->dccptxh_slab == NULL) - goto out_free_slab_name; -out: - return hist; -out_free_slab_name: - kfree(slab_name); -out_free_hist: - kfree(hist); - hist = NULL; - goto out; -} - -EXPORT_SYMBOL_GPL(dccp_tx_hist_new); - -void dccp_tx_hist_delete(struct dccp_tx_hist *hist) -{ - const char* name = kmem_cache_name(hist->dccptxh_slab); - - kmem_cache_destroy(hist->dccptxh_slab); - kfree(name); - kfree(hist); -} - -EXPORT_SYMBOL_GPL(dccp_tx_hist_delete); - -struct dccp_tx_hist_entry * - dccp_tx_hist_find_entry(const struct list_head *list, const u64 seq) -{ - struct dccp_tx_hist_entry *packet = NULL, *entry; - - list_for_each_entry(entry, list, dccphtx_node) - if (entry->dccphtx_seqno == seq) { - packet = entry; - break; - } - - return packet; -} - -EXPORT_SYMBOL_GPL(dccp_tx_hist_find_entry); - -int dccp_rx_hist_find_entry(const struct list_head *list, const u64 seq, - u8 *ccval) -{ - struct dccp_rx_hist_entry *packet = NULL, *entry; - - list_for_each_entry(entry, list, dccphrx_node) - if (entry->dccphrx_seqno == seq) { - packet = entry; - break; - } - - if (packet) - *ccval = packet->dccphrx_ccval; - - return packet != NULL; -} - -EXPORT_SYMBOL_GPL(dccp_rx_hist_find_entry); - -void dccp_tx_hist_purge_older(struct dccp_tx_hist *hist, - struct list_head *list, - struct dccp_tx_hist_entry *packet) -{ - struct dccp_tx_hist_entry *next; - - list_for_each_entry_safe_continue(packet, next, list, dccphtx_node) { - list_del_init(&packet->dccphtx_node); - dccp_tx_hist_entry_delete(hist, packet); + list_for_each_entry_safe(entry, next, list, dccphrx_node) { + list_del_init(&entry->dccphrx_node); + kmem_cache_free(hist->dccprxh_slab, entry); } } -EXPORT_SYMBOL_GPL(dccp_tx_hist_purge_older); +EXPORT_SYMBOL_GPL(dccp_rx_hist_purge); -void dccp_tx_hist_purge(struct dccp_tx_hist *hist, struct list_head *list) -{ - struct dccp_tx_hist_entry *entry, *next; - - list_for_each_entry_safe(entry, next, list, dccphtx_node) { - list_del_init(&entry->dccphtx_node); - dccp_tx_hist_entry_delete(hist, entry); - } -} - -EXPORT_SYMBOL_GPL(dccp_tx_hist_purge); MODULE_AUTHOR("Ian McDonald , " "Arnaldo Carvalho de Melo "); From 9e8efc824098c241a1cde81f5558d222cb6f9369 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:24:57 -0200 Subject: [PATCH 26/32] [DCCP] ccid3: BUG-FIX - conversion errors This fixes conversion errors which arose by not properly type-casting from u32 to __u64. Fixed by explicitly casting each type which is not __u64, or by performing operation after assignment. The patch further adds missing debug information to track the current value of X_recv. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 54 +++++++++++++++++++++++------------------- 1 file changed, 30 insertions(+), 24 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index d23af7b475a0..6bc77ff20908 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -123,18 +123,19 @@ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) if (hctx->ccid3hctx_p > 0) { - hctx->ccid3hctx_x = min_t(u64, hctx->ccid3hctx_x_calc << 6, - hctx->ccid3hctx_x_recv * 2 ); - hctx->ccid3hctx_x = max_t(u64, hctx->ccid3hctx_x, - (hctx->ccid3hctx_s << 6)/TFRC_T_MBI); + hctx->ccid3hctx_x = min(((__u64)hctx->ccid3hctx_x_calc) << 6, + hctx->ccid3hctx_x_recv * 2 ); + hctx->ccid3hctx_x = max(hctx->ccid3hctx_x, + (((__u64)hctx->ccid3hctx_s) << 6) / + TFRC_T_MBI); } else if (timeval_delta(now, &hctx->ccid3hctx_t_ld) - (suseconds_t)hctx->ccid3hctx_rtt >= 0 ) { - hctx->ccid3hctx_x = max(2 * min(hctx->ccid3hctx_x, - hctx->ccid3hctx_x_recv), - scaled_div(hctx->ccid3hctx_s << 6, - hctx->ccid3hctx_rtt )); + hctx->ccid3hctx_x = + max(2 * min(hctx->ccid3hctx_x, hctx->ccid3hctx_x_recv), + scaled_div(((__u64)hctx->ccid3hctx_s) << 6, + hctx->ccid3hctx_rtt ) ); hctx->ccid3hctx_t_ld = *now; } @@ -207,8 +208,9 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) switch (hctx->ccid3hctx_state) { case TFRC_SSTATE_NO_FBACK: /* RFC 3448, 4.4: Halve send rate directly */ - hctx->ccid3hctx_x = max_t(u32, hctx->ccid3hctx_x / 2, - (hctx->ccid3hctx_s << 6)/TFRC_T_MBI); + hctx->ccid3hctx_x = max(hctx->ccid3hctx_x / 2, + (((__u64)hctx->ccid3hctx_s) << 6) / + TFRC_T_MBI); ccid3_pr_debug("%s(%p, state=%s), updated tx rate to %u " "bytes/s\n", dccp_role(sk), sk, @@ -226,7 +228,7 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) */ if (!hctx->ccid3hctx_idle || (hctx->ccid3hctx_x_recv >= 4 * - scaled_div(hctx->ccid3hctx_s << 6, hctx->ccid3hctx_rtt))) { + scaled_div(((__u64)hctx->ccid3hctx_s) << 6, hctx->ccid3hctx_rtt))) { struct timeval now; ccid3_pr_debug("%s(%p, state=%s), not idle\n", @@ -249,15 +251,16 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) hctx->ccid3hctx_x_calc > (hctx->ccid3hctx_x_recv >> 5)) { hctx->ccid3hctx_x_recv = - max_t(u64, hctx->ccid3hctx_x_recv / 2, - (hctx->ccid3hctx_s << 6) / + max(hctx->ccid3hctx_x_recv / 2, + (((__u64)hctx->ccid3hctx_s) << 6) / (2*TFRC_T_MBI)); if (hctx->ccid3hctx_p == 0) dccp_timestamp(sk, &now); - } else - hctx->ccid3hctx_x_recv = hctx->ccid3hctx_x_calc << 4; - + } else { + hctx->ccid3hctx_x_recv = hctx->ccid3hctx_x_calc; + hctx->ccid3hctx_x_recv <<= 4; + } /* Now recalculate X [RFC 3448, 4.3, step (4)] */ ccid3_hc_tx_update_x(sk, &now); } @@ -320,7 +323,8 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) /* Set initial sending rate X/s to 1pps (X is scaled by 2^6) */ ccid3_hc_tx_update_s(hctx, skb->len); - hctx->ccid3hctx_x = hctx->ccid3hctx_s << 6; + hctx->ccid3hctx_x = hctx->ccid3hctx_s; + hctx->ccid3hctx_x <<= 6; /* First timeout, according to [RFC 3448, 4.2], is 1 second */ hctx->ccid3hctx_t_ipi = USEC_PER_SEC; @@ -421,7 +425,8 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) } /* Update receive rate in units of 64 * bytes/second */ - hctx->ccid3hctx_x_recv = opt_recv->ccid3or_receive_rate << 6; + hctx->ccid3hctx_x_recv = opt_recv->ccid3or_receive_rate; + hctx->ccid3hctx_x_recv <<= 6; /* Update loss event rate */ pinv = opt_recv->ccid3or_loss_event_rate; @@ -460,15 +465,15 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) * Larger Initial Windows [RFC 4342, sec. 5] * We deviate in that we use `s' instead of `MSS'. */ - u16 w_init = min( 4 * hctx->ccid3hctx_s, - max(2 * hctx->ccid3hctx_s, 4380)); + __u64 w_init = min( 4 * hctx->ccid3hctx_s, + max(2 * hctx->ccid3hctx_s, 4380)); hctx->ccid3hctx_rtt = r_sample; - hctx->ccid3hctx_x = scaled_div(w_init<< 6, r_sample); + hctx->ccid3hctx_x = scaled_div(w_init << 6, r_sample); hctx->ccid3hctx_t_ld = now; ccid3_update_send_time(hctx); - ccid3_pr_debug("%s(%p), s=%u, w_init=%u, " + ccid3_pr_debug("%s(%p), s=%u, w_init=%llu, " "R_sample=%ldus, X=%u\n", dccp_role(sk), sk, hctx->ccid3hctx_s, w_init, r_sample, (unsigned)(hctx->ccid3hctx_x >> 6)); @@ -487,11 +492,12 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) ccid3_hc_tx_update_x(sk, &now); ccid3_pr_debug("%s(%p), RTT=%uus (sample=%ldus), s=%u, " - "p=%u, X_calc=%u, X=%u\n", dccp_role(sk), + "p=%u, X_calc=%u, X_recv=%u, X=%u\n", dccp_role(sk), sk, hctx->ccid3hctx_rtt, r_sample, hctx->ccid3hctx_s, hctx->ccid3hctx_p, hctx->ccid3hctx_x_calc, - (unsigned)(hctx->ccid3hctx_x >> 6)); + (unsigned)(hctx->ccid3hctx_x_recv >> 6), + (unsigned)(hctx->ccid3hctx_x >> 6) ); } /* unschedule no feedback timer */ From 1fba78b6cba14bd37fdb12c5367f1e4d58ff2e0f Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sun, 10 Dec 2006 15:39:29 -0200 Subject: [PATCH 27/32] [DCCP] ccid3: Fixup some type conversions related to rtts Spotted by David Miller when compiling on sparc64, I reproduced it here on parisc64, that are the only platforms to define __kernel_suseconds_t as an 'int', all the others, x86_64 and x86 included typedef it as a 'long', but from the definition of suseconds_t it should just be an 'int' on platforms where it is >= 32bits, it would not require all the castings from suseconds_t to (int) when printking variables of this type, that are not needed on parisc64 and sparc64. Signed-off-by: Arnaldo Carvalho de Melo --- net/dccp/ccids/ccid3.c | 18 +++++++++--------- net/dccp/ccids/ccid3.h | 6 +++--- 2 files changed, 12 insertions(+), 12 deletions(-) diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 6bc77ff20908..5e817e2f1ebc 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -399,7 +399,7 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) struct timeval now; unsigned long t_nfb; u32 pinv; - long r_sample, t_elapsed; + suseconds_t r_sample, t_elapsed; BUG_ON(hctx == NULL); @@ -446,8 +446,8 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) DCCP_BUG_ON(r_sample < 0); if (unlikely(r_sample <= t_elapsed)) - DCCP_WARN("WARNING: r_sample=%ldus <= t_elapsed=%ldus\n", - r_sample, t_elapsed); + DCCP_WARN("WARNING: r_sample=%dus <= t_elapsed=%dus\n", + (int)r_sample, (int)t_elapsed); else r_sample -= t_elapsed; CCID3_RTT_SANITY_CHECK(r_sample); @@ -474,8 +474,8 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) ccid3_update_send_time(hctx); ccid3_pr_debug("%s(%p), s=%u, w_init=%llu, " - "R_sample=%ldus, X=%u\n", dccp_role(sk), - sk, hctx->ccid3hctx_s, w_init, r_sample, + "R_sample=%dus, X=%u\n", dccp_role(sk), + sk, hctx->ccid3hctx_s, w_init, (int)r_sample, (unsigned)(hctx->ccid3hctx_x >> 6)); ccid3_hc_tx_set_state(sk, TFRC_SSTATE_FBACK); @@ -491,9 +491,9 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) hctx->ccid3hctx_p); ccid3_hc_tx_update_x(sk, &now); - ccid3_pr_debug("%s(%p), RTT=%uus (sample=%ldus), s=%u, " + ccid3_pr_debug("%s(%p), RTT=%uus (sample=%dus), s=%u, " "p=%u, X_calc=%u, X_recv=%u, X=%u\n", dccp_role(sk), - sk, hctx->ccid3hctx_rtt, r_sample, + sk, hctx->ccid3hctx_rtt, (int)r_sample, hctx->ccid3hctx_s, hctx->ccid3hctx_p, hctx->ccid3hctx_x_calc, (unsigned)(hctx->ccid3hctx_x_recv >> 6), @@ -821,8 +821,8 @@ found: DCCP_BUG_ON(delta < 0); rtt = delta * 4 / interval; - ccid3_pr_debug("%s(%p), approximated RTT to %ldus\n", - dccp_role(sk), sk, rtt); + ccid3_pr_debug("%s(%p), approximated RTT to %dus\n", + dccp_role(sk), sk, (int)rtt); /* * Determine the length of the first loss interval via inverse lookup. diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index 3fa0f69ed320..3e3a9cc69b10 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -52,12 +52,12 @@ #define TFRC_T_MBI 64 /* What we think is a reasonable upper limit on RTT values */ -#define CCID3_SANE_RTT_MAX (4 * USEC_PER_SEC) +#define CCID3_SANE_RTT_MAX ((suseconds_t)(4 * USEC_PER_SEC)) #define CCID3_RTT_SANITY_CHECK(rtt) do { \ if (rtt > CCID3_SANE_RTT_MAX) { \ - DCCP_CRIT("RTT (%ld) too large, substituting %ld", \ - rtt, CCID3_SANE_RTT_MAX); \ + DCCP_CRIT("RTT (%d) too large, substituting %d", \ + (int)rtt, (int)CCID3_SANE_RTT_MAX); \ rtt = CCID3_SANE_RTT_MAX; \ } } while (0) From 8109b02b5397ed52a32c116163a62a34f4768b26 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sun, 10 Dec 2006 16:01:18 -0200 Subject: [PATCH 28/32] [DCCP]: Whitespace cleanups That accumulated over the last months hackaton, shame on me for not using git-apply whitespace helping hand, will do that from now on. Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 26 +++++----- net/dccp/ackvec.c | 4 +- net/dccp/ccids/ccid2.c | 12 ++--- net/dccp/ccids/ccid3.c | 111 +++++++++++++++++++++++------------------ net/dccp/ccids/ccid3.h | 22 ++++---- net/dccp/feat.c | 6 +-- net/dccp/input.c | 26 +++++----- net/dccp/ipv4.c | 26 +++++----- net/dccp/ipv6.c | 24 ++++----- net/dccp/minisocks.c | 2 +- net/dccp/output.c | 7 ++- net/dccp/proto.c | 6 +-- net/dccp/timer.c | 14 +++--- 13 files changed, 149 insertions(+), 137 deletions(-) diff --git a/include/linux/dccp.h b/include/linux/dccp.h index ed6cc8962d87..1cb054bd93f2 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -176,20 +176,20 @@ enum { }; /* DCCP features (RFC 4340 section 6.4) */ - enum { - DCCPF_RESERVED = 0, - DCCPF_CCID = 1, +enum { + DCCPF_RESERVED = 0, + DCCPF_CCID = 1, DCCPF_SHORT_SEQNOS = 2, /* XXX: not yet implemented */ - DCCPF_SEQUENCE_WINDOW = 3, + DCCPF_SEQUENCE_WINDOW = 3, DCCPF_ECN_INCAPABLE = 4, /* XXX: not yet implemented */ - DCCPF_ACK_RATIO = 5, - DCCPF_SEND_ACK_VECTOR = 6, - DCCPF_SEND_NDP_COUNT = 7, + 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 */ - /* 10-127 reserved */ - DCCPF_MIN_CCID_SPECIFIC = 128, - DCCPF_MAX_CCID_SPECIFIC = 255, + /* 10-127 reserved */ + DCCPF_MIN_CCID_SPECIFIC = 128, + DCCPF_MAX_CCID_SPECIFIC = 255, }; /* this structure is argument to DCCP_SOCKOPT_CHANGE_X */ @@ -427,7 +427,7 @@ struct dccp_service_list { }; #define DCCP_SERVICE_INVALID_VALUE htonl((__u32)-1) -#define DCCP_SERVICE_CODE_IS_ABSENT 0 +#define DCCP_SERVICE_CODE_IS_ABSENT 0 static inline int dccp_list_has_service(const struct dccp_service_list *sl, const __be32 service) @@ -436,7 +436,7 @@ static inline int dccp_list_has_service(const struct dccp_service_list *sl, u32 i = sl->dccpsl_nr; while (i--) if (sl->dccpsl_list[i] == service) - return 1; + return 1; } return 0; } @@ -511,7 +511,7 @@ struct dccp_sock { __u8 dccps_hc_tx_insert_options:1; struct timer_list dccps_xmit_timer; }; - + static inline struct dccp_sock *dccp_sk(const struct sock *sk) { return (struct dccp_sock *)sk; diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c index 1f4727ddbdbf..a086c6312d3b 100644 --- a/net/dccp/ackvec.c +++ b/net/dccp/ackvec.c @@ -223,7 +223,7 @@ static inline int dccp_ackvec_set_buf_head_state(struct dccp_ackvec *av, gap = -new_head; } new_head += DCCP_MAX_ACKVEC_LEN; - } + } av->dccpav_buf_head = new_head; @@ -336,7 +336,7 @@ out_duplicate: void dccp_ackvector_print(const u64 ackno, const unsigned char *vector, int len) { dccp_pr_debug_cat("ACK vector len=%d, ackno=%llu |", len, - (unsigned long long)ackno); + (unsigned long long)ackno); while (len--) { const u8 state = (*vector & DCCP_ACKVEC_STATE_MASK) >> 6; diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c index 2555be8f4790..fd38b05d6f79 100644 --- a/net/dccp/ccids/ccid2.c +++ b/net/dccp/ccids/ccid2.c @@ -351,7 +351,7 @@ static void ccid2_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) while (seqp != hctx->ccid2hctx_seqh) { ccid2_pr_debug("out seq=%llu acked=%d time=%lu\n", - (unsigned long long)seqp->ccid2s_seq, + (unsigned long long)seqp->ccid2s_seq, seqp->ccid2s_acked, seqp->ccid2s_sent); seqp = seqp->ccid2s_next; } @@ -473,7 +473,7 @@ static inline void ccid2_new_ack(struct sock *sk, /* first measurement */ if (hctx->ccid2hctx_srtt == -1) { ccid2_pr_debug("R: %lu Time=%lu seq=%llu\n", - r, jiffies, + r, jiffies, (unsigned long long)seqp->ccid2s_seq); ccid2_change_srtt(hctx, r); hctx->ccid2hctx_rttvar = r >> 1; @@ -518,8 +518,8 @@ static inline void ccid2_new_ack(struct sock *sk, 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_srtt, hctx->ccid2hctx_rttvar, + hctx->ccid2hctx_rto, HZ, r); hctx->ccid2hctx_sent = 0; } @@ -667,9 +667,9 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) /* new packet received or marked */ if (state != DCCP_ACKVEC_STATE_NOT_RECEIVED && !seqp->ccid2s_acked) { - if (state == + if (state == DCCP_ACKVEC_STATE_ECN_MARKED) { - ccid2_congestion_event(hctx, + ccid2_congestion_event(hctx, seqp); } else ccid2_new_ack(sk, seqp, diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 5e817e2f1ebc..fa6b75372ed7 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -124,18 +124,18 @@ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) if (hctx->ccid3hctx_p > 0) { hctx->ccid3hctx_x = min(((__u64)hctx->ccid3hctx_x_calc) << 6, - hctx->ccid3hctx_x_recv * 2 ); + hctx->ccid3hctx_x_recv * 2); hctx->ccid3hctx_x = max(hctx->ccid3hctx_x, (((__u64)hctx->ccid3hctx_s) << 6) / - TFRC_T_MBI); + TFRC_T_MBI); } else if (timeval_delta(now, &hctx->ccid3hctx_t_ld) - - (suseconds_t)hctx->ccid3hctx_rtt >= 0 ) { + (suseconds_t)hctx->ccid3hctx_rtt >= 0) { hctx->ccid3hctx_x = max(2 * min(hctx->ccid3hctx_x, hctx->ccid3hctx_x_recv), scaled_div(((__u64)hctx->ccid3hctx_s) << 6, - hctx->ccid3hctx_rtt ) ); + hctx->ccid3hctx_rtt)); hctx->ccid3hctx_t_ld = *now; } @@ -144,8 +144,8 @@ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) } /* - * Track the mean packet size `s' (cf. RFC 4342, 5.3 and RFC 3448, 4.1) - * @len: DCCP packet payload size in bytes + * Track the mean packet size `s' (cf. RFC 4342, 5.3 and RFC 3448, 4.1) + * @len: DCCP packet payload size in bytes */ static inline void ccid3_hc_tx_update_s(struct ccid3_hc_tx_sock *hctx, int len) { @@ -163,8 +163,8 @@ static inline void ccid3_hc_tx_update_s(struct ccid3_hc_tx_sock *hctx, int len) } /* - * Update Window Counter using the algorithm from [RFC 4342, 8.1]. - * The algorithm is not applicable if RTT < 4 microseconds. + * Update Window Counter using the algorithm from [RFC 4342, 8.1]. + * The algorithm is not applicable if RTT < 4 microseconds. */ static inline void ccid3_hc_tx_update_win_count(struct ccid3_hc_tx_sock *hctx, struct timeval *now) @@ -228,12 +228,13 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) */ if (!hctx->ccid3hctx_idle || (hctx->ccid3hctx_x_recv >= 4 * - scaled_div(((__u64)hctx->ccid3hctx_s) << 6, hctx->ccid3hctx_rtt))) { + scaled_div(((__u64)hctx->ccid3hctx_s) << 6, + hctx->ccid3hctx_rtt))) { struct timeval now; ccid3_pr_debug("%s(%p, state=%s), not idle\n", dccp_role(sk), sk, - ccid3_tx_state_name(hctx->ccid3hctx_state)); + ccid3_tx_state_name(hctx->ccid3hctx_state)); /* * Modify the cached value of X_recv [RFC 3448, 4.4] @@ -248,12 +249,13 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) BUG_ON(hctx->ccid3hctx_p && !hctx->ccid3hctx_x_calc); if (hctx->ccid3hctx_p == 0 || - hctx->ccid3hctx_x_calc > (hctx->ccid3hctx_x_recv >> 5)) { + (hctx->ccid3hctx_x_calc > + (hctx->ccid3hctx_x_recv >> 5))) { hctx->ccid3hctx_x_recv = max(hctx->ccid3hctx_x_recv / 2, (((__u64)hctx->ccid3hctx_s) << 6) / - (2*TFRC_T_MBI)); + (2 * TFRC_T_MBI)); if (hctx->ccid3hctx_p == 0) dccp_timestamp(sk, &now); @@ -316,7 +318,8 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) switch (hctx->ccid3hctx_state) { case TFRC_SSTATE_NO_SENT: sk_reset_timer(sk, &hctx->ccid3hctx_no_feedback_timer, - jiffies + usecs_to_jiffies(TFRC_INITIAL_TIMEOUT)); + (jiffies + + usecs_to_jiffies(TFRC_INITIAL_TIMEOUT))); hctx->ccid3hctx_last_win_count = 0; hctx->ccid3hctx_t_last_win_count = now; ccid3_hc_tx_set_state(sk, TFRC_SSTATE_NO_FBACK); @@ -338,7 +341,7 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) case TFRC_SSTATE_FBACK: delay = timeval_delta(&hctx->ccid3hctx_t_nom, &now); /* - * Scheduling of packet transmissions [RFC 3448, 4.6] + * Scheduling of packet transmissions [RFC 3448, 4.6] * * if (t_now > t_nom - delta) * // send the packet now @@ -365,7 +368,8 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) return 0; } -static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) +static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, + unsigned int len) { struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); struct timeval now; @@ -415,12 +419,12 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) case TFRC_SSTATE_FBACK: /* get packet from history to look up t_recvdata */ packet = dccp_tx_hist_find_entry(&hctx->ccid3hctx_hist, - DCCP_SKB_CB(skb)->dccpd_ack_seq); + DCCP_SKB_CB(skb)->dccpd_ack_seq); if (unlikely(packet == NULL)) { DCCP_WARN("%s(%p), seqno %llu(%s) doesn't exist " "in history!\n", dccp_role(sk), sk, (unsigned long long)DCCP_SKB_CB(skb)->dccpd_ack_seq, - dccp_packet_name(DCCP_SKB_CB(skb)->dccpd_type)); + dccp_packet_name(DCCP_SKB_CB(skb)->dccpd_type)); return; } @@ -433,13 +437,13 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) if (pinv == ~0U || pinv == 0) /* see RFC 4342, 8.5 */ hctx->ccid3hctx_p = 0; else /* can not exceed 100% */ - hctx->ccid3hctx_p = 1000000 / pinv; + hctx->ccid3hctx_p = 1000000 / pinv; dccp_timestamp(sk, &now); /* * Calculate new round trip sample as per [RFC 3448, 4.3] by - * R_sample = (now - t_recvdata) - t_elapsed + * R_sample = (now - t_recvdata) - t_elapsed */ r_sample = timeval_delta(&now, &packet->dccphtx_tstamp); t_elapsed = dp->dccps_options_received.dccpor_elapsed_time * 10; @@ -465,7 +469,7 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) * Larger Initial Windows [RFC 4342, sec. 5] * We deviate in that we use `s' instead of `MSS'. */ - __u64 w_init = min( 4 * hctx->ccid3hctx_s, + __u64 w_init = min(4 * hctx->ccid3hctx_s, max(2 * hctx->ccid3hctx_s, 4380)); hctx->ccid3hctx_rtt = r_sample; hctx->ccid3hctx_x = scaled_div(w_init << 6, r_sample); @@ -475,13 +479,14 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) ccid3_pr_debug("%s(%p), s=%u, w_init=%llu, " "R_sample=%dus, X=%u\n", dccp_role(sk), - sk, hctx->ccid3hctx_s, w_init, (int)r_sample, + sk, hctx->ccid3hctx_s, w_init, + (int)r_sample, (unsigned)(hctx->ccid3hctx_x >> 6)); ccid3_hc_tx_set_state(sk, TFRC_SSTATE_FBACK); } else { hctx->ccid3hctx_rtt = (9 * hctx->ccid3hctx_rtt + - (u32)r_sample ) / 10; + (u32)r_sample) / 10; /* Update sending rate (step 4 of [RFC 3448, 4.3]) */ if (hctx->ccid3hctx_p > 0) @@ -492,12 +497,13 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) ccid3_hc_tx_update_x(sk, &now); ccid3_pr_debug("%s(%p), RTT=%uus (sample=%dus), s=%u, " - "p=%u, X_calc=%u, X_recv=%u, X=%u\n", dccp_role(sk), + "p=%u, X_calc=%u, X_recv=%u, X=%u\n", + dccp_role(sk), sk, hctx->ccid3hctx_rtt, (int)r_sample, hctx->ccid3hctx_s, hctx->ccid3hctx_p, hctx->ccid3hctx_x_calc, (unsigned)(hctx->ccid3hctx_x_recv >> 6), - (unsigned)(hctx->ccid3hctx_x >> 6) ); + (unsigned)(hctx->ccid3hctx_x >> 6)); } /* unschedule no feedback timer */ @@ -507,20 +513,20 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) dccp_tx_hist_purge_older(ccid3_tx_hist, &hctx->ccid3hctx_hist, packet); /* - * As we have calculated new ipi, delta, t_nom it is possible that - * we now can send a packet, so wake up dccp_wait_for_ccid + * As we have calculated new ipi, delta, t_nom it is possible + * that we now can send a packet, so wake up dccp_wait_for_ccid */ sk->sk_write_space(sk); /* * Update timeout interval for the nofeedback timer. * We use a configuration option to increase the lower bound. - * This can help avoid triggering the nofeedback timer too often - * ('spinning') on LANs with small RTTs. + * This can help avoid triggering the nofeedback timer too + * often ('spinning') on LANs with small RTTs. */ hctx->ccid3hctx_t_rto = max_t(u32, 4 * hctx->ccid3hctx_rtt, CONFIG_IP_DCCP_CCID3_RTO * - (USEC_PER_SEC/1000) ); + (USEC_PER_SEC/1000)); /* * Schedule no feedback timer to expire in * max(t_RTO, 2 * s/X) = max(t_RTO, 2 * t_ipi) @@ -528,7 +534,8 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) t_nfb = max(hctx->ccid3hctx_t_rto, 2 * hctx->ccid3hctx_t_ipi); ccid3_pr_debug("%s(%p), Scheduled no feedback timer to " - "expire in %lu jiffies (%luus)\n", dccp_role(sk), + "expire in %lu jiffies (%luus)\n", + dccp_role(sk), sk, usecs_to_jiffies(t_nfb), t_nfb); sk_reset_timer(sk, &hctx->ccid3hctx_no_feedback_timer, @@ -538,7 +545,9 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) hctx->ccid3hctx_idle = 1; break; case TFRC_SSTATE_NO_SENT: - /* XXX when implementing bidirectional rx/tx check this again */ + /* + * XXX when implementing bidirectional rx/tx check this again + */ DCCP_WARN("Illegal ACK received - no packet sent\n"); /* fall through */ case TFRC_SSTATE_TERM: /* ignore feedback when closing */ @@ -575,7 +584,8 @@ static int ccid3_hc_tx_parse_options(struct sock *sk, unsigned char option, dccp_role(sk), sk, len); rc = -EINVAL; } else { - opt_recv->ccid3or_loss_event_rate = ntohl(*(__be32 *)value); + opt_recv->ccid3or_loss_event_rate = + ntohl(*(__be32 *)value); ccid3_pr_debug("%s(%p), LOSS_EVENT_RATE=%u\n", dccp_role(sk), sk, opt_recv->ccid3or_loss_event_rate); @@ -596,7 +606,8 @@ static int ccid3_hc_tx_parse_options(struct sock *sk, unsigned char option, dccp_role(sk), sk, len); rc = -EINVAL; } else { - opt_recv->ccid3or_receive_rate = ntohl(*(__be32 *)value); + opt_recv->ccid3or_receive_rate = + ntohl(*(__be32 *)value); ccid3_pr_debug("%s(%p), RECEIVE_RATE=%u\n", dccp_role(sk), sk, opt_recv->ccid3or_receive_rate); @@ -616,7 +627,8 @@ static int ccid3_hc_tx_init(struct ccid *ccid, struct sock *sk) hctx->ccid3hctx_state = TFRC_SSTATE_NO_SENT; INIT_LIST_HEAD(&hctx->ccid3hctx_hist); - hctx->ccid3hctx_no_feedback_timer.function = ccid3_hc_tx_no_feedback_timer; + hctx->ccid3hctx_no_feedback_timer.function = + ccid3_hc_tx_no_feedback_timer; hctx->ccid3hctx_no_feedback_timer.data = (unsigned long)sk; init_timer(&hctx->ccid3hctx_no_feedback_timer); @@ -754,9 +766,9 @@ static int ccid3_hc_rx_insert_options(struct sock *sk, struct sk_buff *skb) hcrx->ccid3hcrx_elapsed_time)) || dccp_insert_option_timestamp(sk, skb) || dccp_insert_option(sk, skb, TFRC_OPT_LOSS_EVENT_RATE, - &pinv, sizeof(pinv)) || + &pinv, sizeof(pinv)) || dccp_insert_option(sk, skb, TFRC_OPT_RECEIVE_RATE, - &x_recv, sizeof(x_recv))) + &x_recv, sizeof(x_recv))) return -1; return 0; @@ -827,9 +839,9 @@ found: /* * Determine the length of the first loss interval via inverse lookup. * Assume that X_recv can be computed by the throughput equation - * s - * X_recv = -------- - * R * fval + * s + * X_recv = -------- + * R * fval * Find some p such that f(p) = fval; return 1/p [RFC 3448, 6.3.1]. */ if (rtt == 0) { /* would result in divide-by-zero */ @@ -860,7 +872,7 @@ found: if (p == 0) return ~0; else - return 1000000 / p; + return 1000000 / p; } static void ccid3_hc_rx_update_li(struct sock *sk, u64 seq_loss, u8 win_loss) @@ -914,7 +926,8 @@ static int ccid3_hc_rx_detect_loss(struct sock *sk, struct dccp_rx_hist_entry *packet) { struct ccid3_hc_rx_sock *hcrx = ccid3_hc_rx_sk(sk); - struct dccp_rx_hist_entry *rx_hist = dccp_rx_hist_head(&hcrx->ccid3hcrx_hist); + struct dccp_rx_hist_entry *rx_hist = + dccp_rx_hist_head(&hcrx->ccid3hcrx_hist); u64 seqno = packet->dccphrx_seqno; u64 tmp_seqno; int loss = 0; @@ -942,7 +955,7 @@ static int ccid3_hc_rx_detect_loss(struct sock *sk, dccp_inc_seqno(&tmp_seqno); while (dccp_rx_hist_find_entry(&hcrx->ccid3hcrx_hist, tmp_seqno, &ccval)) { - hcrx->ccid3hcrx_seqno_nonloss = tmp_seqno; + hcrx->ccid3hcrx_seqno_nonloss = tmp_seqno; hcrx->ccid3hcrx_ccval_nonloss = ccval; dccp_inc_seqno(&tmp_seqno); } @@ -1044,8 +1057,8 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) break; dccp_timestamp(sk, &now); - if (timeval_delta(&now, &hcrx->ccid3hcrx_tstamp_last_ack) - - (suseconds_t)hcrx->ccid3hcrx_rtt >= 0) { + if ((timeval_delta(&now, &hcrx->ccid3hcrx_tstamp_last_ack) - + (suseconds_t)hcrx->ccid3hcrx_rtt) >= 0) { hcrx->ccid3hcrx_tstamp_last_ack = now; ccid3_hc_rx_send_feedback(sk); } @@ -1118,9 +1131,9 @@ static void ccid3_hc_rx_get_info(struct sock *sk, struct tcp_info *info) BUG_ON(hcrx == NULL); - info->tcpi_ca_state = hcrx->ccid3hcrx_state; - info->tcpi_options |= TCPI_OPT_TIMESTAMPS; - info->tcpi_rcv_rtt = hcrx->ccid3hcrx_rtt; + info->tcpi_ca_state = hcrx->ccid3hcrx_state; + info->tcpi_options |= TCPI_OPT_TIMESTAMPS; + info->tcpi_rcv_rtt = hcrx->ccid3hcrx_rtt; } static void ccid3_hc_tx_get_info(struct sock *sk, struct tcp_info *info) @@ -1212,7 +1225,7 @@ static struct ccid_operations ccid3 = { .ccid_hc_rx_getsockopt = ccid3_hc_rx_getsockopt, .ccid_hc_tx_getsockopt = ccid3_hc_tx_getsockopt, }; - + #ifdef CONFIG_IP_DCCP_CCID3_DEBUG module_param(ccid3_debug, int, 0444); MODULE_PARM_DESC(ccid3_debug, "Enable debug messages"); @@ -1235,7 +1248,7 @@ static __init int ccid3_module_init(void) goto out_free_tx; rc = ccid_register(&ccid3); - if (rc != 0) + if (rc != 0) goto out_free_loss_interval_history; out: return rc; diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index 3e3a9cc69b10..15776a88c090 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -77,7 +77,7 @@ struct ccid3_options_received { /* TFRC sender states */ enum ccid3_hc_tx_states { - TFRC_SSTATE_NO_SENT = 1, + TFRC_SSTATE_NO_SENT = 1, TFRC_SSTATE_NO_FBACK, TFRC_SSTATE_FBACK, TFRC_SSTATE_TERM, @@ -96,7 +96,7 @@ enum ccid3_hc_tx_states { * @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 + * with last_win_count value sent * @ccid3hctx_no_feedback_timer - Handle to no feedback timer * @ccid3hctx_idle - Flag indicating that sender is idling * @ccid3hctx_t_ld - Time last doubled during slow start @@ -115,7 +115,7 @@ struct ccid3_hc_tx_sock { #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; + enum ccid3_hc_tx_states ccid3hctx_state:8; u8 ccid3hctx_last_win_count; u8 ccid3hctx_idle; struct timeval ccid3hctx_t_last_win_count; @@ -129,7 +129,7 @@ struct ccid3_hc_tx_sock { /* TFRC receiver states */ enum ccid3_hc_rx_states { - TFRC_RSTATE_NO_DATA = 1, + TFRC_RSTATE_NO_DATA = 1, TFRC_RSTATE_DATA, TFRC_RSTATE_TERM = 127, }; @@ -157,18 +157,18 @@ struct ccid3_hc_rx_sock { #define ccid3hcrx_x_recv ccid3hcrx_tfrc.tfrcrx_x_recv #define ccid3hcrx_rtt ccid3hcrx_tfrc.tfrcrx_rtt #define ccid3hcrx_p ccid3hcrx_tfrc.tfrcrx_p - u64 ccid3hcrx_seqno_nonloss:48, + u64 ccid3hcrx_seqno_nonloss:48, ccid3hcrx_ccval_nonloss:4, ccid3hcrx_ccval_last_counter:4; enum ccid3_hc_rx_states ccid3hcrx_state:8; - u32 ccid3hcrx_bytes_recv; - struct timeval ccid3hcrx_tstamp_last_feedback; - struct timeval ccid3hcrx_tstamp_last_ack; + u32 ccid3hcrx_bytes_recv; + struct timeval ccid3hcrx_tstamp_last_feedback; + struct timeval ccid3hcrx_tstamp_last_ack; struct list_head ccid3hcrx_hist; struct list_head ccid3hcrx_li_hist; - u16 ccid3hcrx_s; - u32 ccid3hcrx_pinv; - u32 ccid3hcrx_elapsed_time; + u16 ccid3hcrx_s; + u32 ccid3hcrx_pinv; + u32 ccid3hcrx_elapsed_time; }; static inline struct ccid3_hc_tx_sock *ccid3_hc_tx_sk(const struct sock *sk) diff --git a/net/dccp/feat.c b/net/dccp/feat.c index 4dc487f27a1f..95b6927ec653 100644 --- a/net/dccp/feat.c +++ b/net/dccp/feat.c @@ -329,7 +329,7 @@ static void dccp_feat_empty_confirm(struct dccp_minisock *dmsk, switch (type) { case DCCPO_CHANGE_L: opt->dccpop_type = DCCPO_CONFIRM_R; break; case DCCPO_CHANGE_R: opt->dccpop_type = DCCPO_CONFIRM_L; break; - default: DCCP_WARN("invalid type %d\n", type); return; + default: DCCP_WARN("invalid type %d\n", type); return; } opt->dccpop_feat = feature; @@ -427,7 +427,7 @@ int dccp_feat_confirm_recv(struct sock *sk, u8 type, u8 feature, switch (type) { case DCCPO_CONFIRM_L: t = DCCPO_CHANGE_R; break; case DCCPO_CONFIRM_R: t = DCCPO_CHANGE_L; break; - default: DCCP_WARN("invalid type %d\n", type); + default: DCCP_WARN("invalid type %d\n", type); return 1; } @@ -610,7 +610,7 @@ const char *dccp_feat_typename(const u8 type) case DCCPO_CHANGE_R: return("ChangeR"); case DCCPO_CONFIRM_R: return("ConfirmR"); /* the following case must not appear in feature negotation */ - default: dccp_pr_debug("unknown type %d [BUG!]\n", type); + default: dccp_pr_debug("unknown type %d [BUG!]\n", type); } return NULL; } diff --git a/net/dccp/input.c b/net/dccp/input.c index 4a3279cd684c..565bc80557ce 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -1,6 +1,6 @@ /* * net/dccp/input.c - * + * * An implementation of the DCCP protocol * Arnaldo Carvalho de Melo * @@ -82,7 +82,7 @@ static int dccp_check_seqno(struct sock *sk, struct sk_buff *skb) * Otherwise, * Drop packet and return */ - if (dh->dccph_type == DCCP_PKT_SYNC || + if (dh->dccph_type == DCCP_PKT_SYNC || dh->dccph_type == DCCP_PKT_SYNCACK) { if (between48(DCCP_SKB_CB(skb)->dccpd_ack_seq, dp->dccps_awl, dp->dccps_awh) && @@ -185,8 +185,8 @@ static int __dccp_rcv_established(struct sock *sk, struct sk_buff *skb, dccp_rcv_close(sk, skb); return 0; case DCCP_PKT_REQUEST: - /* Step 7 - * or (S.is_server and P.type == Response) + /* Step 7 + * or (S.is_server and P.type == Response) * or (S.is_client and P.type == Request) * or (S.state >= OPEN and P.type == Request * and P.seqno >= S.OSR) @@ -274,7 +274,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, const struct dccp_hdr *dh, const unsigned len) { - /* + /* * Step 4: Prepare sequence numbers in REQUEST * If S.state == REQUEST, * If (P.type == Response or P.type == Reset) @@ -342,7 +342,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, * from the Response * / * S.state := PARTOPEN * Set PARTOPEN timer - * Continue with S.state == PARTOPEN + * Continue with S.state == PARTOPEN * / * Step 12 will send the Ack completing the * three-way handshake * / */ @@ -373,7 +373,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, */ __kfree_skb(skb); return 0; - } + } dccp_send_ack(sk); return -1; } @@ -381,7 +381,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, out_invalid_packet: /* dccp_v4_do_rcv will send a reset */ DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_PACKET_ERROR; - return 1; + return 1; } static int dccp_rcv_respond_partopen_state_process(struct sock *sk, @@ -488,11 +488,11 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, if (dcb->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_event_ack_recv(sk, skb); - if (dccp_msk(sk)->dccpms_send_ack_vector && + if (dccp_msk(sk)->dccpms_send_ack_vector && dccp_ackvec_add(dp->dccps_hc_rx_ackvec, sk, - DCCP_SKB_CB(skb)->dccpd_seq, - DCCP_ACKVEC_STATE_RECEIVED)) - goto discard; + DCCP_SKB_CB(skb)->dccpd_seq, + DCCP_ACKVEC_STATE_RECEIVED)) + goto discard; /* XXX see the comments in dccp_rcv_established about this */ if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER) @@ -580,7 +580,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, } } - if (!queued) { + if (!queued) { discard: __kfree_skb(skb); } diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index ff81679c9f17..90c74b4adb73 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -157,7 +157,7 @@ static inline void dccp_do_pmtu_discovery(struct sock *sk, /* We don't check in the destentry if pmtu discovery is forbidden * on this route. We just assume that no packet_to_big packets * are send back when pmtu discovery is not active. - * There is a small race when the user changes this flag in the + * There is a small race when the user changes this flag in the * route, but I think that's acceptable. */ if ((dst = __sk_dst_check(sk, 0)) == NULL) @@ -467,7 +467,7 @@ static struct dst_entry* dccp_v4_route_skb(struct sock *sk, .uli_u = { .ports = { .sport = dccp_hdr(skb)->dccph_dport, .dport = dccp_hdr(skb)->dccph_sport } - } + } }; security_skb_classify_flow(skb, &fl); @@ -595,7 +595,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) struct inet_request_sock *ireq; struct request_sock *req; struct dccp_request_sock *dreq; - const __be32 service = dccp_hdr_request(skb)->dccph_req_service; + const __be32 service = dccp_hdr_request(skb)->dccph_req_service; struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); __u8 reset_code = DCCP_RESET_CODE_TOO_BUSY; @@ -609,7 +609,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (dccp_bad_service_code(sk, service)) { reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE; goto drop; - } + } /* * TW buckets are converted to open requests without * limitations, they conserve resources and peer is @@ -644,7 +644,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) ireq->rmt_addr = skb->nh.iph->saddr; ireq->opt = NULL; - /* + /* * Step 3: Process LISTEN state * * Set S.ISR, S.GSR, S.SWL, S.SWH from packet or Init Cookie @@ -846,15 +846,15 @@ static int dccp_v4_rcv(struct sk_buff *skb) } /* Step 2: - * Look up flow ID in table and get corresponding socket */ + * Look up flow ID in table and get corresponding socket */ sk = __inet_lookup(&dccp_hashinfo, skb->nh.iph->saddr, dh->dccph_sport, skb->nh.iph->daddr, dh->dccph_dport, inet_iif(skb)); - /* + /* * Step 2: - * If no socket ... + * If no socket ... */ if (sk == NULL) { dccp_pr_debug("failed to look up flow ID in table and " @@ -862,9 +862,9 @@ static int dccp_v4_rcv(struct sk_buff *skb) goto no_dccp_socket; } - /* + /* * Step 2: - * ... or S.state == TIMEWAIT, + * ... or S.state == TIMEWAIT, * Generate Reset(No Connection) unless P.type == Reset * Drop packet and return */ @@ -876,8 +876,8 @@ static int dccp_v4_rcv(struct sk_buff *skb) /* * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage - * o if MinCsCov = 0, only packets with CsCov = 0 are accepted - * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov + * o if MinCsCov = 0, only packets with CsCov = 0 are accepted + * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov */ min_cov = dccp_sk(sk)->dccps_pcrlen; if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) { @@ -900,7 +900,7 @@ no_dccp_socket: goto discard_it; /* * Step 2: - * If no socket ... + * If no socket ... * Generate Reset(No Connection) unless P.type == Reset * Drop packet and return */ diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index c7aaa2574f52..6b91a9dd0411 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -77,7 +77,7 @@ static inline void dccp_v6_send_check(struct sock *sk, int unused_value, } static inline __u32 secure_dccpv6_sequence_number(__be32 *saddr, __be32 *daddr, - __be16 sport, __be16 dport ) + __be16 sport, __be16 dport ) { return secure_tcpv6_sequence_number(saddr, daddr, sport, dport); } @@ -329,7 +329,7 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) skb = alloc_skb(dccp_v6_ctl_socket->sk->sk_prot->max_header, GFP_ATOMIC); if (skb == NULL) - return; + return; skb_reserve(skb, dccp_v6_ctl_socket->sk->sk_prot->max_header); @@ -353,7 +353,7 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) dccp_csum_outgoing(skb); dh->dccph_checksum = dccp_v6_csum_finish(skb, &rxskb->nh.ipv6h->saddr, - &rxskb->nh.ipv6h->daddr); + &rxskb->nh.ipv6h->daddr); memset(&fl, 0, sizeof(fl)); ipv6_addr_copy(&fl.fl6_dst, &rxskb->nh.ipv6h->saddr); @@ -424,7 +424,7 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) struct dccp_request_sock *dreq; struct inet6_request_sock *ireq6; struct ipv6_pinfo *np = inet6_sk(sk); - const __be32 service = dccp_hdr_request(skb)->dccph_req_service; + const __be32 service = dccp_hdr_request(skb)->dccph_req_service; struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); __u8 reset_code = DCCP_RESET_CODE_TOO_BUSY; @@ -437,7 +437,7 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (dccp_bad_service_code(sk, service)) { reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE; goto drop; - } + } /* * There are no SYN attacks on IPv6, yet... */ @@ -787,7 +787,7 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) * otherwise we just shortcircuit this and continue with * the new socket.. */ - if (nsk != sk) { + if (nsk != sk) { if (dccp_child_process(sk, nsk, skb)) goto reset; if (opt_skb != NULL) @@ -843,14 +843,14 @@ static int dccp_v6_rcv(struct sk_buff **pskb) DCCP_SKB_CB(skb)->dccpd_ack_seq = dccp_hdr_ack_seq(skb); /* Step 2: - * Look up flow ID in table and get corresponding socket */ + * Look up flow ID in table and get corresponding socket */ sk = __inet6_lookup(&dccp_hashinfo, &skb->nh.ipv6h->saddr, dh->dccph_sport, &skb->nh.ipv6h->daddr, ntohs(dh->dccph_dport), inet6_iif(skb)); /* * Step 2: - * If no socket ... + * If no socket ... */ if (sk == NULL) { dccp_pr_debug("failed to look up flow ID in table and " @@ -860,7 +860,7 @@ static int dccp_v6_rcv(struct sk_buff **pskb) /* * Step 2: - * ... or S.state == TIMEWAIT, + * ... or S.state == TIMEWAIT, * Generate Reset(No Connection) unless P.type == Reset * Drop packet and return */ @@ -872,8 +872,8 @@ static int dccp_v6_rcv(struct sk_buff **pskb) /* * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage - * o if MinCsCov = 0, only packets with CsCov = 0 are accepted - * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov + * o if MinCsCov = 0, only packets with CsCov = 0 are accepted + * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov */ min_cov = dccp_sk(sk)->dccps_pcrlen; if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) { @@ -893,7 +893,7 @@ no_dccp_socket: goto discard_it; /* * Step 2: - * If no socket ... + * If no socket ... * Generate Reset(No Connection) unless P.type == Reset * Drop packet and return */ diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 4c9e26775f72..6656bb497c7b 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -182,7 +182,7 @@ out_free: EXPORT_SYMBOL_GPL(dccp_create_openreq_child); -/* +/* * Process an incoming packet for RESPOND sockets represented * as an request_sock. */ diff --git a/net/dccp/output.c b/net/dccp/output.c index b4df12b24161..824569659083 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -1,6 +1,6 @@ /* * net/dccp/output.c - * + * * An implementation of the DCCP protocol * Arnaldo Carvalho de Melo * @@ -338,7 +338,6 @@ EXPORT_SYMBOL_GPL(dccp_make_response); static struct sk_buff *dccp_make_reset(struct sock *sk, struct dst_entry *dst, const enum dccp_reset_codes code) - { struct dccp_hdr *dh; struct dccp_sock *dp = dccp_sk(sk); @@ -419,14 +418,14 @@ static inline void dccp_connect_init(struct sock *sk) dccp_sync_mss(sk, dst_mtu(dst)); - /* + /* * SWL and AWL are initially adjusted so that they are not less than * the initial Sequence Numbers received and sent, respectively: * SWL := max(GSR + 1 - floor(W/4), ISR), * AWL := max(GSS - W' + 1, ISS). * These adjustments MUST be applied only at the beginning of the * connection. - */ + */ dccp_update_gss(sk, dp->dccps_iss); dccp_set_seqno(&dp->dccps_awl, max48(dp->dccps_awl, dp->dccps_iss)); diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 5ec47d9ee447..63b3fa20e14b 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -196,7 +196,7 @@ int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized) sk, GFP_KERNEL); dp->dccps_hc_tx_ccid = ccid_hc_tx_new(dmsk->dccpms_tx_ccid, sk, GFP_KERNEL); - if (unlikely(dp->dccps_hc_rx_ccid == NULL || + if (unlikely(dp->dccps_hc_rx_ccid == NULL || dp->dccps_hc_tx_ccid == NULL)) { ccid_hc_rx_delete(dp->dccps_hc_rx_ccid, sk); ccid_hc_tx_delete(dp->dccps_hc_tx_ccid, sk); @@ -390,7 +390,7 @@ static int dccp_setsockopt_service(struct sock *sk, const __be32 service, struct dccp_sock *dp = dccp_sk(sk); struct dccp_service_list *sl = NULL; - if (service == DCCP_SERVICE_INVALID_VALUE || + if (service == DCCP_SERVICE_INVALID_VALUE || optlen > DCCP_SERVICE_LIST_MAX_LEN * sizeof(u32)) return -EINVAL; @@ -830,7 +830,7 @@ EXPORT_SYMBOL_GPL(inet_dccp_listen); static const unsigned char dccp_new_state[] = { /* current state: new state: action: */ [0] = DCCP_CLOSED, - [DCCP_OPEN] = DCCP_CLOSING | DCCP_ACTION_FIN, + [DCCP_OPEN] = DCCP_CLOSING | DCCP_ACTION_FIN, [DCCP_REQUESTING] = DCCP_CLOSED, [DCCP_PARTOPEN] = DCCP_CLOSING | DCCP_ACTION_FIN, [DCCP_LISTEN] = DCCP_CLOSED, diff --git a/net/dccp/timer.c b/net/dccp/timer.c index e8f519e7f481..e5348f369c60 100644 --- a/net/dccp/timer.c +++ b/net/dccp/timer.c @@ -1,6 +1,6 @@ /* * net/dccp/timer.c - * + * * An implementation of the DCCP protocol * Arnaldo Carvalho de Melo * @@ -102,13 +102,13 @@ static void dccp_retransmit_timer(struct sock *sk) * sk->sk_send_head has to have one skb with * DCCP_SKB_CB(skb)->dccpd_type set to one of the retransmittable DCCP * packet types. The only packets eligible for retransmission are: - * -- Requests in client-REQUEST state (sec. 8.1.1) - * -- Acks in client-PARTOPEN state (sec. 8.1.5) - * -- CloseReq in server-CLOSEREQ state (sec. 8.3) - * -- Close in node-CLOSING state (sec. 8.3) */ + * -- Requests in client-REQUEST state (sec. 8.1.1) + * -- Acks in client-PARTOPEN state (sec. 8.1.5) + * -- CloseReq in server-CLOSEREQ state (sec. 8.3) + * -- Close in node-CLOSING state (sec. 8.3) */ BUG_TRAP(sk->sk_send_head != NULL); - /* + /* * More than than 4MSL (8 minutes) has passed, a RESET(aborted) was * sent, no need to retransmit, this sock is dead. */ @@ -200,7 +200,7 @@ static void dccp_keepalive_timer(unsigned long data) /* Only process if socket is not in use. */ bh_lock_sock(sk); if (sock_owned_by_user(sk)) { - /* Try again later. */ + /* Try again later. */ inet_csk_reset_keepalive_timer(sk, HZ / 20); goto out; } From f654c854d1d4e0aca5389ace541647237eb7f753 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sun, 10 Dec 2006 13:46:45 -0800 Subject: [PATCH 29/32] [HAMRADIO]: Fix baycom_epp.c compile failure. Fix foobar in 15b1c0e822f578306332d4f4c449250db5c5dceb and e8cc49bb0fdb9e18a99e6780073d1400ba2b0d1f patch series. Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- drivers/net/hamradio/baycom_epp.c | 2 +- include/net/ax25.h | 2 ++ net/ax25/ax25_addr.c | 2 +- 3 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/hamradio/baycom_epp.c b/drivers/net/hamradio/baycom_epp.c index 8a83db0fb3b7..153b6dc80af4 100644 --- a/drivers/net/hamradio/baycom_epp.c +++ b/drivers/net/hamradio/baycom_epp.c @@ -1177,7 +1177,7 @@ static void baycom_probe(struct net_device *dev) dev->mtu = AX25_DEF_PACLEN; /* eth_mtu is the default */ dev->addr_len = AX25_ADDR_LEN; /* sizeof an ax.25 address */ memcpy(dev->broadcast, &ax25_bcast, AX25_ADDR_LEN); - memcpy(dev->dev_addr, &ax25_nocall, AX25_ADDR_LEN); + memcpy(dev->dev_addr, &null_ax25_address, AX25_ADDR_LEN); dev->tx_queue_len = 16; /* New style flags */ diff --git a/include/net/ax25.h b/include/net/ax25.h index e1d116f11923..14b72d868f03 100644 --- a/include/net/ax25.h +++ b/include/net/ax25.h @@ -285,6 +285,8 @@ extern struct sock *ax25_make_new(struct sock *, struct ax25_dev *); extern const ax25_address ax25_bcast; extern const ax25_address ax25_defaddr; extern const ax25_address null_ax25_address; +extern char *ax2asc(char *buf, const ax25_address *); +extern void asc2ax(ax25_address *addr, const char *callsign); extern int ax25cmp(const ax25_address *, const ax25_address *); extern int ax25digicmp(const ax25_digi *, const ax25_digi *); extern const unsigned char *ax25_addr_parse(const unsigned char *, int, diff --git a/net/ax25/ax25_addr.c b/net/ax25/ax25_addr.c index 21a0616152fc..97a49c79c605 100644 --- a/net/ax25/ax25_addr.c +++ b/net/ax25/ax25_addr.c @@ -83,7 +83,7 @@ EXPORT_SYMBOL(ax2asc); */ void asc2ax(ax25_address *addr, const char *callsign) { - char *s; + const char *s; int n; for (s = callsign, n = 0; n < 6; n++) { From 89eaeb09ba01632511f69ce8e14c8d2abd2dbe9e Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Sun, 10 Dec 2006 22:35:24 -0800 Subject: [PATCH 30/32] [IPVS]: Use msleep_interruptable() instead of ssleep() aka msleep() Dean Manners notices that when an IPVS synchonisation daemons are started the system load slowly climbs up to 1. This seems to be related to the call to ssleep(1) (aka msleep(1000) in the main loop. Replacing this with a call to msleep_interruptable() seems to make the problem go away. Though I'm not sure that it is correct. This is the second edition of this patch, which replaces ssleep() in the main loop for both the master and backup threads, as well as some thread synchronisation code. The latter is just for thorougness as it shouldn't be causing any problems. Signed-Off-By: Simon Horman Signed-off-by: David S. Miller --- net/ipv4/ipvs/ip_vs_sync.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/ipv4/ipvs/ip_vs_sync.c b/net/ipv4/ipvs/ip_vs_sync.c index 91a075edd68e..7461eeb3a3dd 100644 --- a/net/ipv4/ipvs/ip_vs_sync.c +++ b/net/ipv4/ipvs/ip_vs_sync.c @@ -657,7 +657,7 @@ static void sync_master_loop(void) if (stop_master_sync) break; - ssleep(1); + msleep_interruptible(1000); } /* clean up the sync_buff queue */ @@ -714,7 +714,7 @@ static void sync_backup_loop(void) if (stop_backup_sync) break; - ssleep(1); + msleep_interruptible(1000); } /* release the sending multicast socket */ @@ -826,7 +826,7 @@ static int fork_sync_thread(void *startup) if ((pid = kernel_thread(sync_thread, startup, 0)) < 0) { IP_VS_ERR("could not create sync_thread due to %d... " "retrying.\n", pid); - ssleep(1); + msleep_interruptible(1000); goto repeat; } @@ -860,7 +860,7 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) if ((pid = kernel_thread(fork_sync_thread, &startup, 0)) < 0) { IP_VS_ERR("could not create fork_sync_thread due to %d... " "retrying.\n", pid); - ssleep(1); + msleep_interruptible(1000); goto repeat; } From 37004af3aaf28f39981754a05cf74361c0834a7a Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Sun, 10 Dec 2006 22:36:18 -0800 Subject: [PATCH 31/32] [IPVS]: Make ip_vs_sync.c <= 80col wide. Signed-off-by: Simon Horman Signed-off-by: David S. Miller --- net/ipv4/ipvs/ip_vs_sync.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/net/ipv4/ipvs/ip_vs_sync.c b/net/ipv4/ipvs/ip_vs_sync.c index 7461eeb3a3dd..7ea2d981a932 100644 --- a/net/ipv4/ipvs/ip_vs_sync.c +++ b/net/ipv4/ipvs/ip_vs_sync.c @@ -849,10 +849,12 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) ip_vs_sync_state |= state; if (state == IP_VS_STATE_MASTER) { - strlcpy(ip_vs_master_mcast_ifn, mcast_ifn, sizeof(ip_vs_master_mcast_ifn)); + strlcpy(ip_vs_master_mcast_ifn, mcast_ifn, + sizeof(ip_vs_master_mcast_ifn)); ip_vs_master_syncid = syncid; } else { - strlcpy(ip_vs_backup_mcast_ifn, mcast_ifn, sizeof(ip_vs_backup_mcast_ifn)); + strlcpy(ip_vs_backup_mcast_ifn, mcast_ifn, + sizeof(ip_vs_backup_mcast_ifn)); ip_vs_backup_syncid = syncid; } @@ -880,7 +882,8 @@ int stop_sync_thread(int state) IP_VS_DBG(7, "%s: pid %d\n", __FUNCTION__, current->pid); IP_VS_INFO("stopping sync thread %d ...\n", - (state == IP_VS_STATE_MASTER) ? sync_master_pid : sync_backup_pid); + (state == IP_VS_STATE_MASTER) ? + sync_master_pid : sync_backup_pid); __set_current_state(TASK_UNINTERRUPTIBLE); add_wait_queue(&stop_sync_wait, &wait); From a49f99ffca57a2eada23b1ac908a405c17859e35 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 11 Dec 2006 17:24:46 -0800 Subject: [PATCH 32/32] [NETPOLL]: Fix local_bh_enable() warning. During boot we get: netconsole: device eth0 not up yet, forcing it e1000: eth0: e1000_watchdog: NIC Link is Up 100 Mbps Full Duplex WARNING (!__warned) at kernel/softirq.c:137 local_bh_enable() Call Trace: [] local_bh_enable+0x41/0xa3 [] netpoll_send_skb+0x116/0x144 [] netpoll_send_udp+0x263/0x271 [] write_msg+0x42/0x5e [] __call_console_drivers+0x5f/0x70 [] _call_console_drivers+0x6d/0x71 [] release_console_sem+0x148/0x1ec [] register_console+0x1b1/0x1ba [] init_netconsole+0x54/0x68 [] init+0x152/0x308 [] _spin_unlock_irq+0x14/0x30 [] schedule_tail+0x43/0x9f [] child_rip+0xa/0x12 Herbert sayeth: Normally networking isn't invoked with interrupts turned off, but I suppose we don't have a choice here. This is unique being a place where you can get called with BH on, off, or IRQs off. Given that this is only used for printk, the easiest solution is probably just to disable local IRQs instead of BH. Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- net/core/netpoll.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 156d7fd81fec..a01abdd2d3ea 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -243,7 +243,9 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) /* don't get messages out of order, and no recursion */ if (skb_queue_len(&npinfo->txq) == 0 && npinfo->poll_owner != smp_processor_id()) { - local_bh_disable(); /* Where's netif_tx_trylock_bh()? */ + unsigned long flags; + + local_irq_save(flags); if (netif_tx_trylock(dev)) { /* try until next clock tick */ for (tries = jiffies_to_usecs(1)/USEC_PER_POLL; @@ -261,7 +263,7 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) } netif_tx_unlock(dev); } - local_bh_enable(); + local_irq_restore(flags); } if (status != NETDEV_TX_OK) {