openwrt/upstream - upstream openwrt

	Commit message (Expand)	Author	Age	Files	Lines
*	map: rename type to maptype (FS#3287)	Remi NGUYEN VAN	2020-08-19	1	-13/+18
*	map: add a legacymap option	Remi NGUYEN VAN	2020-08-15	1	-7/+5
*	map: don't set default firewall zone to wan	Hans Dedecker	2019-05-26	1	-22/+26
*	map: drop default encaplimit value	Hans Dedecker	2018-09-19	1	-1/+1
*	map: make tunnel encapsulation limit support configurable (FS#1501)	Hans Dedecker	2018-06-04	1	-7/+9
*	map: use nested json data object to store map-e fmrs parameters	Hans Dedecker	2017-08-29	1	-6/+10
*	map: add ealen as configurable uci parameter	Hans Dedecker	2017-08-24	1	-0/+1
*	map: delete map-t device when tearing down map interface	Hans Dedecker	2017-01-12	1	-1/+11
*	map: add sleep work-around for lw4o6 race-condition	Steven Barth	2015-06-09	1	-0/+3
*	map: shorten autogenerated sub-interface names to account for limits	Steven Barth	2015-04-30	1	-2/+2
*	map: Fix white space errors	John Crispin	2015-02-03	1	-3/+3
*	map: export calculated ruleset to /tmp	Steven Barth	2015-01-13	1	-0/+3
*	map: fix portsets starting with 0 and use regular NAT for 1:1 MAP	Steven Barth	2014-10-02	1	-13/+22
*	map: add support for map-t if available	Steven Barth	2014-06-04	1	-1/+47
*	Initial support for MAP-E and Lightweight 4over6 protocol	Steven Barth	2014-05-22	1	-0/+160

From b8fc328668a74e1314a19266755a54abd875e5a6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet <edumazet@google.com> Date: Sun, 29 Jul 2012 20:52:21 +0000 Subject: [PATCH] codel: refine one condition to avoid a nul rec_inv_sqrt commit 2359a47671fc4fb0fe5e9945f76c2cb10792c0f8 upstream. One condition before codel_Newton_step() was not good if we never left the dropping state for a flow. As a result rec_inv_sqrt was 0, instead of the ~0 initial value. codel control law was then set to a very aggressive mode, dropping many packets before reaching 'target' and recovering from this problem. To keep codel_vars_init() as efficient as possible, refine the condition to make sure rec_inv_sqrt initial value is correct Many thanks to Anton Mich for discovering the issue and suggesting a fix. Reported-by: Anton Mich <lp2s1h@gmail.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> --- include/net/codel.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) --- a/include/net/codel.h +++ b/include/net/codel.h @@ -305,6 +305,8 @@ static struct sk_buff *codel_dequeue(str } } } else if (drop) { + u32 delta; + if (params->ecn && INET_ECN_set_ce(skb)) { stats->ecn_mark++; } else { @@ -320,9 +322,11 @@ static struct sk_buff *codel_dequeue(str * assume that the drop rate that controlled the queue on the * last cycle is a good starting point to control it now. */ - if (codel_time_before(now - vars->drop_next, + delta = vars->count - vars->lastcount; + if (delta > 1 && + codel_time_before(now - vars->drop_next, 16 * params->interval)) { - vars->count = (vars->count - vars->lastcount) | 1; + vars->count = delta; /* we dont care if rec_inv_sqrt approximation * is not very precise : * Next Newton steps will correct it quadratically.