Repository files (latest commit first)

Filename	Latest commit message	Latest commit date
Gavin McCullagh 2072c228c9 [TCP]: use non-delayed ACK for congestion control RTT ... When a delayed ACK representing two packets arrives, there are two RTT samples available, one for each packet. The first (in order of seq number) will be artificially long due to the delay waiting for the second packet, the second will trigger the ACK and so will not itself be delayed. According to rfc1323, the SRTT used for RTO calculation should use the first rtt, so receivers echo the timestamp from the first packet in the delayed ack. For congestion control however, it seems measuring delayed ack delay is not desirable as it varies independently of congestion. The patch below causes seq_rtt and last_ackt to be updated with any available later packet rtts which should have less (and hopefully zero) delack delay. The rtt value then gets passed to ca_ops->pkts_acked(). Where TCP_CONG_RTT_STAMP was set, effort was made to supress RTTs from within a TSO chunk (!fully_acked), using only the final ACK (which includes any TSO delay) to generate RTTs. This patch removes these checks so RTTs are passed for each ACK to ca_ops->pkts_acked(). For non-delay based congestion control (cubic, h-tcp), rtt is sometimes used for rtt-scaling. In shortening the RTT, this may make them a little less aggressive. Delay-based schemes (eg vegas, veno, illinois) should get a cleaner, more accurate congestion signal, particularly for small cwnds. The congestion control module can potentially also filter out bad RTTs due to the delayed ack alarm by looking at the associated cnt which (where delayed acking is in use) should probably be 1 if the alarm went off or greater if the ACK was triggered by a packet. Signed-off-by: Gavin McCullagh <[email protected]> Acked-by: Ilpo J�rvinen <[email protected]> Signed-off-by: David S. Miller <[email protected]>		2007-12-29 19:11:21 -08:00
..
ipvs	[IPVS]: Fix sched registration race when checking for name collision.	2007-12-05 05:37:27 -08:00
netfilter	[NETFILTER]: nf_conntrack_ipv4: fix module parameter compatibility	2007-12-26 19:36:33 -08:00
af_inet.c	[NET]: Forget the zero_it argument of sk_alloc()	2007-11-01 00:39:31 -07:00
ah4.c	cleanup asm/scatterlist.h includes	2007-11-02 08:47:06 +01:00
arp.c	[IPV4] ARP: Remove not used code	2007-12-19 23:38:11 -08:00
cipso_ipv4.c	[NetLabel]: correct usage of RCU locking	2007-10-26 04:29:08 -07:00
datagram.c
devinet.c	[IPV4]: Swap the ifa allocation with the"ipv4_devconf_setall" call	2007-12-11 02:45:25 -08:00
esp4.c	[IPv4] ESP: Discard dummy packets introduced in rfc4303	2007-12-11 02:45:26 -08:00
fib_frontend.c	[IPV4]: OOPS with NETLINK_FIB_LOOKUP netlink socket	2007-12-21 02:01:53 -08:00
fib_hash.c	[NET]: Make core networking code use seq_open_private	2007-10-10 16:55:33 -07:00
fib_lookup.h
fib_rules.c	[INET]: Small possible memory leak in FIB rules	2007-11-10 22:12:03 -08:00
fib_semantics.c	[NETLINK]: Introduce nested and byteorder flag to netlink attribute	2007-10-10 16:49:16 -07:00
fib_trie.c	remove asm/bitops.h includes	2007-10-19 11:53:41 -07:00
icmp.c	[INET]: Unexport icmpmsg_statistics	2007-10-26 04:06:08 -07:00
igmp.c	[NET]: Make core networking code use seq_open_private	2007-10-10 16:55:33 -07:00
inet_connection_sock.c	[INET]: Justification for local port range robustness.	2007-10-18 22:00:17 -07:00
inet_diag.c	[INET]: Fix inet_diag dead-lock regression	2007-12-03 15:51:25 +11:00
inet_fragment.c	[INET]: Consolidate frag queues freeing	2007-10-17 19:48:26 -07:00
inet_hashtables.c	[INET]: Remove per bucket rwlock in tcp/dccp ehash table.	2007-11-07 04:15:11 -08:00
inet_lro.c	[LRO]: fix lro_gen_skb() alignment	2007-12-05 05:37:32 -08:00
inet_timewait_sock.c	[INET]: Remove per bucket rwlock in tcp/dccp ehash table.	2007-11-07 04:15:11 -08:00
inetpeer.c	[INET]: Use list_head-s in inetpeer.c	2007-11-12 21:27:28 -08:00
ip_forward.c	[IPV4]: Uninline netfilter okfns	2007-10-15 12:26:35 -07:00
ip_fragment.c	[NET]: Fix uninitialised variable in ip_frag_reasm()	2007-10-17 21:37:22 -07:00
ip_gre.c	[IPV4] ip_gre: set mac_header correctly in receive path	2007-12-20 00:10:33 -08:00
ip_input.c	[IPV4]: Uninline netfilter okfns	2007-10-15 12:26:35 -07:00
ip_options.c
ip_output.c	[IPV4]: Consolidate the ip cork destruction in ip_output.c	2007-11-07 04:08:25 -08:00
ip_sockglue.c	[IPV4]: Clean the ip_sockglue.c from some ugly ifdefs	2007-11-07 04:08:55 -08:00
ipcomp.c	[IPSEC]: Fix crypto_alloc_comp error checking	2007-11-07 04:15:03 -08:00
ipconfig.c	[IPV4] Fix ip=dhcp regression	2007-12-28 13:39:11 -08:00
ipip.c	[NET]: Treat the sign of the result of skb_headroom() consistently	2007-10-23 21:27:55 -07:00
ipmr.c	[NET]: Make core networking code use seq_open_private	2007-10-10 16:55:33 -07:00
Kconfig	typo fixes	2007-10-20 01:34:40 +02:00
Makefile	[INET]: Collect frag queues management objects together	2007-10-15 12:26:39 -07:00
netfilter.c	[NETFILTER]: Replace sk_buff ** with sk_buff *	2007-10-15 12:26:29 -07:00
proc.c	[NET]: Define infrastructure to keep 'inuse' changes in an efficent SMP/NUMA way.	2007-11-07 04:08:57 -08:00
protocol.c
raw.c	[IPV4]: Use the {DEFINE|REF}_PROTO_INUSE infrastructure	2007-11-07 04:08:58 -08:00
route.c	[IPV4]: Remove prototype of ip_rt_advice	2007-12-07 01:07:38 -08:00
syncookies.c
sysctl_net_ipv4.c	[TCP]: Problem bug with sysctl_tcp_congestion_control function	2007-11-19 23:28:21 -08:00
tcp.c	[INET]: Remove per bucket rwlock in tcp/dccp ehash table.	2007-11-07 04:15:11 -08:00
tcp_bic.c	[TCP]: Remove num_acked>0 checks from cong.ctrl mods pkts_acked	2007-10-10 16:47:55 -07:00
tcp_cong.c
tcp_cubic.c	[TCP]: Remove num_acked>0 checks from cong.ctrl mods pkts_acked	2007-10-10 16:47:55 -07:00
tcp_diag.c	[INET]: Let inet_diag and friends autoload	2007-10-22 02:59:54 -07:00
tcp_highspeed.c
tcp_htcp.c
tcp_hybla.c
tcp_illinois.c	[TCP] illinois: Incorrect beta usage	2007-11-30 01:10:55 +11:00
tcp_input.c	[TCP]: use non-delayed ACK for congestion control RTT	2007-12-29 19:11:21 -08:00
tcp_ipv4.c	[IPV4] TCPMD5: Use memmove() instead of memcpy() because we have overlaps.	2007-11-20 17:30:31 -08:00
tcp_lp.c
tcp_minisocks.c	[TCP]: Move sack_ok access to obviously named funcs & cleanup	2007-10-10 16:48:00 -07:00
tcp_output.c	[TCP]: NAGLE_PUSH seems to be a wrong way around	2007-12-05 05:37:31 -08:00
tcp_probe.c	[NET]: Make /proc/net per network namespace	2007-10-10 16:49:06 -07:00
tcp_scalable.c
tcp_timer.c	[TCP]: Move sack_ok access to obviously named funcs & cleanup	2007-10-10 16:48:00 -07:00
tcp_vegas.c	[TCP] vegas: Fix a bug in disabling slow start by gamma parameter.	2007-10-29 22:37:25 -07:00
tcp_vegas.h
tcp_veno.c
tcp_westwood.c
tcp_yeah.c
tunnel4.c	[INET]: Cleanup the xfrm4_tunnel_(un)register	2007-11-10 21:48:54 -08:00
udp.c	[IPV4]: Use the {DEFINE|REF}_PROTO_INUSE infrastructure	2007-11-07 04:08:58 -08:00
udp_impl.h	[UDP]: Randomize port selection.	2007-10-10 16:48:31 -07:00
udplite.c	[IPV4]: Use the {DEFINE|REF}_PROTO_INUSE infrastructure	2007-11-07 04:08:58 -08:00
xfrm4_input.c	[IPSEC]: Rename mode to outer_mode and add inner_mode	2007-10-17 21:35:51 -07:00
xfrm4_mode_beet.c	[IPSEC]: Add missing BEET checks	2007-10-17 21:31:50 -07:00
xfrm4_mode_transport.c	[IPSEC]: Use IPv6 calling convention as the convention for x->mode->output	2007-10-10 16:55:54 -07:00
xfrm4_mode_tunnel.c	[IPSEC]: Add missing BEET checks	2007-10-17 21:31:50 -07:00
xfrm4_output.c	[IPSEC]: Rename mode to outer_mode and add inner_mode	2007-10-17 21:35:51 -07:00
xfrm4_policy.c	[IPSEC]: Rename mode to outer_mode and add inner_mode	2007-10-17 21:35:51 -07:00
xfrm4_state.c	[IPSEC]: Store afinfo pointer in xfrm_mode	2007-10-17 21:33:12 -07:00
xfrm4_tunnel.c	[IPSEC]: Move tunnel parsing for IPv4 out of xfrm4_input	2007-10-17 21:28:53 -07:00