From b9ce204f0a265f819d10c943a607746abb62f245 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Fri, 15 Jun 2007 15:08:43 -0700 Subject: [PATCH 1/3] [TCP]: Congestion control API RTT sampling fix MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 164891aadf1721fca4dce473bb0e0998181537c6 broke RTT sampling of congestion control modules. Inaccurate timestamps could be fed to them without providing any way for them to identify such cases. Previously RTT sampler was called only if FLAG_RETRANS_DATA_ACKED was not set filtering inaccurate timestamps nicely. In addition, the new behavior could give an invalid timestamp (zero) to RTT sampler if only skbs with TCPCB_RETRANS were ACKed. This solves both problems. Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- include/linux/ktime.h | 12 ++++++++++++ include/linux/skbuff.h | 4 ++++ net/ipv4/tcp_illinois.c | 3 +++ net/ipv4/tcp_input.c | 6 +++++- net/ipv4/tcp_lp.c | 3 ++- net/ipv4/tcp_vegas.c | 3 +++ net/ipv4/tcp_veno.c | 3 +++ 7 files changed, 32 insertions(+), 2 deletions(-) diff --git a/include/linux/ktime.h b/include/linux/ktime.h index c762954bda14..2b139f66027f 100644 --- a/include/linux/ktime.h +++ b/include/linux/ktime.h @@ -261,6 +261,18 @@ static inline s64 ktime_to_ns(const ktime_t kt) #endif +/** + * ktime_equal - Compares two ktime_t variables to see if they are equal + * @cmp1: comparable1 + * @cmp2: comparable2 + * + * Compare two ktime_t variables, returns 1 if equal + */ +static inline int ktime_equal(const ktime_t cmp1, const ktime_t cmp2) +{ + return cmp1.tv64 == cmp2.tv64; +} + static inline s64 ktime_to_us(const ktime_t kt) { struct timeval tv = ktime_to_timeval(kt); diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index e7367c74e1bb..6f0b2f7d0010 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1579,6 +1579,10 @@ static inline ktime_t net_timedelta(ktime_t t) return ktime_sub(ktime_get_real(), t); } +static inline ktime_t net_invalid_timestamp(void) +{ + return ktime_set(0, 0); +} extern __sum16 __skb_checksum_complete_head(struct sk_buff *skb, int len); extern __sum16 __skb_checksum_complete(struct sk_buff *skb); diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp_illinois.c index 4adc47c55351..b2b2256d3b84 100644 --- a/net/ipv4/tcp_illinois.c +++ b/net/ipv4/tcp_illinois.c @@ -90,6 +90,9 @@ static void tcp_illinois_acked(struct sock *sk, u32 pkts_acked, ktime_t last) ca->acked = pkts_acked; + if (ktime_equal(last, net_invalid_timestamp())) + return; + rtt = ktime_to_us(net_timedelta(last)); /* ignore bogus values, this prevents wraparound in alpha math */ diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index d6d0f9b6cdc6..aaf6f66677f9 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2409,7 +2409,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) int acked = 0; int prior_packets = tp->packets_out; __s32 seq_rtt = -1; - ktime_t last_ackt = ktime_set(0,0); + ktime_t last_ackt = net_invalid_timestamp(); while ((skb = tcp_write_queue_head(sk)) && skb != tcp_send_head(sk)) { @@ -2487,6 +2487,10 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) tcp_ack_update_rtt(sk, acked, seq_rtt); tcp_ack_packets_out(sk); + /* Is the ACK triggering packet unambiguous? */ + if (acked & FLAG_RETRANS_DATA_ACKED) + last_ackt = net_invalid_timestamp(); + if (ca_ops->pkts_acked) ca_ops->pkts_acked(sk, pkts_acked, last_ackt); } diff --git a/net/ipv4/tcp_lp.c b/net/ipv4/tcp_lp.c index 43294ad9f63e..e49836ce012e 100644 --- a/net/ipv4/tcp_lp.c +++ b/net/ipv4/tcp_lp.c @@ -266,7 +266,8 @@ static void tcp_lp_pkts_acked(struct sock *sk, u32 num_acked, ktime_t last) struct tcp_sock *tp = tcp_sk(sk); struct lp *lp = inet_csk_ca(sk); - tcp_lp_rtt_sample(sk, ktime_to_us(net_timedelta(last))); + if (!ktime_equal(last, net_invalid_timestamp())) + tcp_lp_rtt_sample(sk, ktime_to_us(net_timedelta(last))); /* calc inference */ if (tcp_time_stamp > tp->rx_opt.rcv_tsecr) diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c index 73e19cf7df21..e218a51ceced 100644 --- a/net/ipv4/tcp_vegas.c +++ b/net/ipv4/tcp_vegas.c @@ -117,6 +117,9 @@ void tcp_vegas_pkts_acked(struct sock *sk, u32 cnt, ktime_t last) struct vegas *vegas = inet_csk_ca(sk); u32 vrtt; + if (ktime_equal(last, net_invalid_timestamp())) + return; + /* Never allow zero rtt or baseRTT */ vrtt = ktime_to_us(net_timedelta(last)) + 1; diff --git a/net/ipv4/tcp_veno.c b/net/ipv4/tcp_veno.c index 9edb340f2f95..ec854cc5fad5 100644 --- a/net/ipv4/tcp_veno.c +++ b/net/ipv4/tcp_veno.c @@ -74,6 +74,9 @@ static void tcp_veno_pkts_acked(struct sock *sk, u32 cnt, ktime_t last) struct veno *veno = inet_csk_ca(sk); u32 vrtt; + if (ktime_equal(last, net_invalid_timestamp())) + return; + /* Never allow zero rtt or baseRTT */ vrtt = ktime_to_us(net_timedelta(last)) + 1; From 7769f4064cd9f4fbae09f071578a051aae83e844 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Fri, 15 Jun 2007 15:14:04 -0700 Subject: [PATCH 2/3] [TCP]: Fix logic breakage due to DSACK separation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 6f74651ae626ec672028587bc700538076dfbefb is found guilty of breaking DSACK counting, which should be done only for the SACK block reported by the DSACK instead of every SACK block that is received along with DSACK information. Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index aaf6f66677f9..69f9f1ef3ef6 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -953,7 +953,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ int prior_fackets; u32 lost_retrans = 0; int flag = 0; - int dup_sack = 0; + int found_dup_sack = 0; int cached_fack_count; int i; int first_sack_index; @@ -964,20 +964,20 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ /* Check for D-SACK. */ if (before(ntohl(sp[0].start_seq), TCP_SKB_CB(ack_skb)->ack_seq)) { - dup_sack = 1; + found_dup_sack = 1; tp->rx_opt.sack_ok |= 4; NET_INC_STATS_BH(LINUX_MIB_TCPDSACKRECV); } else if (num_sacks > 1 && !after(ntohl(sp[0].end_seq), ntohl(sp[1].end_seq)) && !before(ntohl(sp[0].start_seq), ntohl(sp[1].start_seq))) { - dup_sack = 1; + found_dup_sack = 1; tp->rx_opt.sack_ok |= 4; NET_INC_STATS_BH(LINUX_MIB_TCPDSACKOFORECV); } /* D-SACK for already forgotten data... * Do dumb counting. */ - if (dup_sack && + if (found_dup_sack && !after(ntohl(sp[0].end_seq), prior_snd_una) && after(ntohl(sp[0].end_seq), tp->undo_marker)) tp->undo_retrans--; @@ -1058,6 +1058,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ __u32 start_seq = ntohl(sp->start_seq); __u32 end_seq = ntohl(sp->end_seq); int fack_count; + int dup_sack = (found_dup_sack && (i == first_sack_index)); skb = cached_skb; fack_count = cached_fack_count; From 16c61add51f2182140637c924687a2aab6b568f9 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Fri, 15 Jun 2007 15:15:43 -0700 Subject: [PATCH 3/3] [RXRPC] net/rxrpc/ar-connection.c: fix NULL dereference This patch fixes a NULL dereference spotted by the Coverity checker. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- net/rxrpc/ar-connection.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/rxrpc/ar-connection.c b/net/rxrpc/ar-connection.c index 43cb3e051ece..482750efc235 100644 --- a/net/rxrpc/ar-connection.c +++ b/net/rxrpc/ar-connection.c @@ -211,7 +211,7 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) conn->header_size = sizeof(struct rxrpc_header); } - _leave(" = %p{%d}", conn, conn->debug_id); + _leave(" = %p{%d}", conn, conn ? conn->debug_id : 0); return conn; }