From 825de27d9e40b3117b29a79d412b7a4b78c5d815 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 27 May 2008 06:33:54 -0700 Subject: dccp ccid-3: Fix "t_ipi explosion" bug The identification of this bug is thanks to Cheng Wei and Tomasz Grobelny. To avoid divide-by-zero, the implementation previously ignored RTTs smaller than 4 microseconds when performing integer division RTT/4. When the RTT reached a value less than 4 microseconds (as observed on loopback), this prevented the Window Counter CCVal value from advancing. As a result, the receiver stopped sending feedback. This in turn caused non-ending expiries of the nofeedback timer at the sender, so that the sending rate was progressively reduced until reaching the minimum of one packet per 64 seconds. The patch fixes this bug by handling integer division more intelligently. Due to consistent use of dccp_sample_rtt(), divide-by-zero-RTT is avoided. Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- net/dccp/ccids/ccid3.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) (limited to 'net/dccp/ccids') diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index cd61dea2eea1..f813077234b7 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -193,22 +193,17 @@ 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. + * As elsewhere, RTT > 0 is assumed by using dccp_sample_rtt(). */ static inline void ccid3_hc_tx_update_win_count(struct ccid3_hc_tx_sock *hctx, ktime_t now) { - u32 quarter_rtts; - - if (unlikely(hctx->ccid3hctx_rtt < 4)) /* avoid divide-by-zero */ - return; - - quarter_rtts = ktime_us_delta(now, hctx->ccid3hctx_t_last_win_count); - quarter_rtts /= hctx->ccid3hctx_rtt / 4; + u32 delta = ktime_us_delta(now, hctx->ccid3hctx_t_last_win_count), + quarter_rtts = (4 * delta) / hctx->ccid3hctx_rtt; 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 += min(quarter_rtts, 5U); hctx->ccid3hctx_last_win_count &= 0xF; /* mod 16 */ } } -- cgit v1.2.3 From 3294f202dc1acd82223e83ef59f272bd87bb06b2 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Wed, 11 Jun 2008 11:19:09 +0100 Subject: dccp ccid-3: Bug-Fix - Zero RTT is possible In commit $(825de27d9e40b3117b29a79d412b7a4b78c5d815) (from 27th May, commit message `dccp ccid-3: Fix "t_ipi explosion" bug'), the CCID-3 window counter computation was fixed to cope with RTTs < 4 microseconds. Such RTTs can be found e.g. when running CCID-3 over loopback. The fix removed a check against RTT < 4, but introduced a divide-by-zero bug. All steady-state RTTs in DCCP are filtered using dccp_sample_rtt(), which ensures non-zero samples. However, a zero RTT is possible on initialisation, when there is no RTT sample from the Request/Response exchange. The fix is to use the fallback-RTT from RFC 4340, 3.4. This is also better than just fixing update_win_count() since it allows other parts of the code to always assume that the RTT is non-zero during the time that the CCID is used. Signed-off-by: Gerrit Renker --- net/dccp/ccids/ccid3.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'net/dccp/ccids') diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index f813077234b7..0474f4c5707a 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -329,8 +329,14 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) hctx->ccid3hctx_x = rfc3390_initial_rate(sk); hctx->ccid3hctx_t_ld = now; } else { - /* Sender does not have RTT sample: X_pps = 1 pkt/sec */ - hctx->ccid3hctx_x = hctx->ccid3hctx_s; + /* + * Sender does not have RTT sample: + * - set fallback RTT (RFC 4340, 3.4) since a RTT value + * is needed in several parts (e.g. window counter); + * - set sending rate X_pps = 1pps as per RFC 3448, 4.2. + */ + hctx->ccid3hctx_rtt = DCCP_FALLBACK_RTT; + hctx->ccid3hctx_x = hctx->ccid3hctx_s; hctx->ccid3hctx_x <<= 6; } ccid3_update_send_interval(hctx); -- cgit v1.2.3 From 1e2f0e5e8376f2a0ada8760fc9d3104e1a81382b Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Wed, 11 Jun 2008 11:19:09 +0100 Subject: dccp: Fix sparse warnings This patch fixes the following sparse warnings: * nested min(max()) expression: net/dccp/ccids/ccid3.c:91:21: warning: symbol '__x' shadows an earlier one net/dccp/ccids/ccid3.c:91:21: warning: symbol '__y' shadows an earlier one * Declaration of function prototypes in .c instead of .h file, resulting in "should it be static?" warnings. * Declared "struct dccpw" static (local to dccp_probe). * Disabled dccp_delayed_ack() - not fully removed due to RFC 4340, 11.3 ("Receivers SHOULD implement delayed acknowledgement timers ..."). * Used a different local variable name to avoid net/dccp/ackvec.c:293:13: warning: symbol 'state' shadows an earlier one net/dccp/ackvec.c:238:33: originally declared here * Removed unused functions `dccp_ackvector_print' and `dccp_ackvec_print'. Signed-off-by: Gerrit Renker --- net/dccp/ccids/ccid3.c | 4 ++-- net/dccp/ccids/lib/tfrc.c | 8 -------- net/dccp/ccids/lib/tfrc.h | 11 +++++++++-- 3 files changed, 11 insertions(+), 12 deletions(-) (limited to 'net/dccp/ccids') diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 0474f4c5707a..a1929f33d703 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -159,8 +159,8 @@ static void ccid3_hc_tx_update_x(struct sock *sk, ktime_t *stamp) } else if (ktime_us_delta(now, hctx->ccid3hctx_t_ld) - (s64)hctx->ccid3hctx_rtt >= 0) { - hctx->ccid3hctx_x = - max(min(2 * hctx->ccid3hctx_x, min_rate), + hctx->ccid3hctx_x = min(2 * hctx->ccid3hctx_x, min_rate); + hctx->ccid3hctx_x = max(hctx->ccid3hctx_x, scaled_div(((__u64)hctx->ccid3hctx_s) << 6, hctx->ccid3hctx_rtt)); hctx->ccid3hctx_t_ld = now; diff --git a/net/dccp/ccids/lib/tfrc.c b/net/dccp/ccids/lib/tfrc.c index d1dfbb8de64c..97ecec0a8e76 100644 --- a/net/dccp/ccids/lib/tfrc.c +++ b/net/dccp/ccids/lib/tfrc.c @@ -14,14 +14,6 @@ module_param(tfrc_debug, bool, 0444); MODULE_PARM_DESC(tfrc_debug, "Enable debug messages"); #endif -extern int tfrc_tx_packet_history_init(void); -extern void tfrc_tx_packet_history_exit(void); -extern int tfrc_rx_packet_history_init(void); -extern void tfrc_rx_packet_history_exit(void); - -extern int tfrc_li_init(void); -extern void tfrc_li_exit(void); - static int __init tfrc_module_init(void) { int rc = tfrc_li_init(); diff --git a/net/dccp/ccids/lib/tfrc.h b/net/dccp/ccids/lib/tfrc.h index 1fb1187bbf1c..ddd8107b927f 100644 --- a/net/dccp/ccids/lib/tfrc.h +++ b/net/dccp/ccids/lib/tfrc.h @@ -58,7 +58,14 @@ static inline u32 tfrc_ewma(const u32 avg, const u32 newval, const u8 weight) return avg ? (weight * avg + (10 - weight) * newval) / 10 : newval; } -extern u32 tfrc_calc_x(u16 s, u32 R, u32 p); -extern u32 tfrc_calc_x_reverse_lookup(u32 fvalue); +extern u32 tfrc_calc_x(u16 s, u32 R, u32 p); +extern u32 tfrc_calc_x_reverse_lookup(u32 fvalue); +extern int tfrc_tx_packet_history_init(void); +extern void tfrc_tx_packet_history_exit(void); +extern int tfrc_rx_packet_history_init(void); +extern void tfrc_rx_packet_history_exit(void); + +extern int tfrc_li_init(void); +extern void tfrc_li_exit(void); #endif /* _TFRC_H_ */ -- cgit v1.2.3 From 1e8a287c79f64226541f5c44aa52d4698bb84cf5 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Wed, 11 Jun 2008 11:19:10 +0100 Subject: dccp ccid-3: TFRC reverse-lookup Bug-Fix This fixes a bug in the reverse lookup of p: given a value f(p), instead of p, the function returned the smallest tabulated value f(p). The smallest tabulated value of 10^6 * f(p) = sqrt(2*p/3) + 12 * sqrt(3*p/8) * (32 * p^3 + p) for p=0.0001 is 8172. Since this value is scaled by 10^6, the outcome of this bug is that a loss of 8172/10^6 = 0.8172% was reported whenever the input was below the table resolution of 0.01%. This means that the value was over 80 times too high, resulting in large spikes of the initial loss interval, thus unnecessarily reducing the throughput. Also corrected the printk format (%u for u32). Signed-off-by: Gerrit Renker --- net/dccp/ccids/lib/tfrc_equation.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net/dccp/ccids') diff --git a/net/dccp/ccids/lib/tfrc_equation.c b/net/dccp/ccids/lib/tfrc_equation.c index e4e64b76c10c..2f20a29cffe4 100644 --- a/net/dccp/ccids/lib/tfrc_equation.c +++ b/net/dccp/ccids/lib/tfrc_equation.c @@ -661,7 +661,7 @@ u32 tfrc_calc_x(u16 s, u32 R, u32 p) EXPORT_SYMBOL_GPL(tfrc_calc_x); -/* +/** * tfrc_calc_x_reverse_lookup - try to find p given f(p) * * @fvalue: function value to match, scaled by 1000000 @@ -676,11 +676,11 @@ u32 tfrc_calc_x_reverse_lookup(u32 fvalue) /* Error cases. */ if (fvalue < tfrc_calc_x_lookup[0][1]) { - DCCP_WARN("fvalue %d smaller than resolution\n", fvalue); - return tfrc_calc_x_lookup[0][1]; + DCCP_WARN("fvalue %u smaller than resolution\n", fvalue); + return TFRC_SMALLEST_P; } if (fvalue > tfrc_calc_x_lookup[TFRC_CALC_X_ARRSIZE - 1][0]) { - DCCP_WARN("fvalue %d exceeds bounds!\n", fvalue); + DCCP_WARN("fvalue %u exceeds bounds!\n", fvalue); return 1000000; } -- cgit v1.2.3 From 7deb0f851003287d7e259bf6b33548b144c0f2d5 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Wed, 11 Jun 2008 11:19:10 +0100 Subject: dccp ccid-3: X truncated due to type conversion This fixes a bug in computing the inter-packet-interval t_ipi = s/X: scaled_div32(a, b) uses u32 for b, but in "scaled_div32(s, X)" the type of the sending rate `X' is u64. Since X is scaled by 2^6, this truncates rates greater than 2^26 Bps (~537 Mbps). Using full 64-bit division now. Signed-off-by: Gerrit Renker --- net/dccp/ccids/lib/tfrc.h | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) (limited to 'net/dccp/ccids') diff --git a/net/dccp/ccids/lib/tfrc.h b/net/dccp/ccids/lib/tfrc.h index ddd8107b927f..ed9857527acf 100644 --- a/net/dccp/ccids/lib/tfrc.h +++ b/net/dccp/ccids/lib/tfrc.h @@ -15,7 +15,7 @@ * (at your option) any later version. */ #include -#include +#include #include "../../dccp.h" /* internal includes that this module exports: */ #include "loss_interval.h" @@ -29,21 +29,19 @@ extern int tfrc_debug; #endif /* integer-arithmetic divisions of type (a * 1000000)/b */ -static inline u64 scaled_div(u64 a, u32 b) +static inline u64 scaled_div(u64 a, u64 b) { BUG_ON(b==0); - a *= 1000000; - do_div(a, b); - return a; + return div64_u64(a * 1000000, b); } -static inline u32 scaled_div32(u64 a, u32 b) +static inline u32 scaled_div32(u64 a, u64 b) { u64 result = scaled_div(a, b); if (result > UINT_MAX) { - DCCP_CRIT("Overflow: a(%llu)/b(%u) > ~0U", - (unsigned long long)a, b); + DCCP_CRIT("Overflow: %llu/%llu > UINT_MAX", + (unsigned long long)a, (unsigned long long)b); return UINT_MAX; } return result; -- cgit v1.2.3