From: Arne Fitzenreiter Date: Mon, 19 May 2014 08:04:06 +0000 (+0200) Subject: kernel: add pie packet sheduler. X-Git-Url: http://git.ipfire.org/?p=people%2Fteissler%2Fipfire-2.x.git;a=commitdiff_plain;h=e898660a30c7fac19e8ba495d6094e7f9a2aa03a kernel: add pie packet sheduler. --- diff --git a/config/kernel/kernel.config.armv5tel-ipfire-kirkwood b/config/kernel/kernel.config.armv5tel-ipfire-kirkwood index 2d9b11521..9bc1d43af 100644 --- a/config/kernel/kernel.config.armv5tel-ipfire-kirkwood +++ b/config/kernel/kernel.config.armv5tel-ipfire-kirkwood @@ -953,6 +953,7 @@ CONFIG_NET_SCH_CHOKE=m CONFIG_NET_SCH_QFQ=m CONFIG_NET_SCH_CODEL=m CONFIG_NET_SCH_FQ_CODEL=m +CONFIG_NET_SCH_PIE=m CONFIG_NET_SCH_INGRESS=m CONFIG_NET_SCH_PLUG=m diff --git a/config/kernel/kernel.config.armv5tel-ipfire-multi b/config/kernel/kernel.config.armv5tel-ipfire-multi index dcc177d3e..127542f9a 100644 --- a/config/kernel/kernel.config.armv5tel-ipfire-multi +++ b/config/kernel/kernel.config.armv5tel-ipfire-multi @@ -1142,6 +1142,7 @@ CONFIG_NET_SCH_CHOKE=m CONFIG_NET_SCH_QFQ=m CONFIG_NET_SCH_CODEL=m CONFIG_NET_SCH_FQ_CODEL=m +CONFIG_NET_SCH_PIE=m CONFIG_NET_SCH_INGRESS=m CONFIG_NET_SCH_PLUG=m diff --git a/config/kernel/kernel.config.armv5tel-ipfire-rpi b/config/kernel/kernel.config.armv5tel-ipfire-rpi index 0c9b3d20e..2c709ccfc 100644 --- a/config/kernel/kernel.config.armv5tel-ipfire-rpi +++ b/config/kernel/kernel.config.armv5tel-ipfire-rpi @@ -888,6 +888,7 @@ CONFIG_NET_SCH_CHOKE=m CONFIG_NET_SCH_QFQ=m CONFIG_NET_SCH_CODEL=m CONFIG_NET_SCH_FQ_CODEL=m +CONFIG_NET_SCH_PIE=m CONFIG_NET_SCH_INGRESS=m CONFIG_NET_SCH_PLUG=m diff --git a/config/kernel/kernel.config.i586-ipfire b/config/kernel/kernel.config.i586-ipfire index 246e436ec..ba63a2976 100644 --- a/config/kernel/kernel.config.i586-ipfire +++ b/config/kernel/kernel.config.i586-ipfire @@ -1,6 +1,6 @@ # # Automatically generated file; DO NOT EDIT. -# Linux/x86 3.10.33 Kernel Configuration +# Linux/x86 3.10.39 Kernel Configuration # # CONFIG_64BIT is not set CONFIG_X86_32=y @@ -913,6 +913,8 @@ CONFIG_NETFILTER_XT_MATCH_HELPER=m CONFIG_NETFILTER_XT_MATCH_HL=m CONFIG_NETFILTER_XT_MATCH_IPRANGE=m # CONFIG_NETFILTER_XT_MATCH_IPVS is not set +CONFIG_NETFILTER_XT_MATCH_LAYER7=m +# CONFIG_NETFILTER_XT_MATCH_LAYER7_DEBUG is not set CONFIG_NETFILTER_XT_MATCH_LENGTH=m CONFIG_NETFILTER_XT_MATCH_LIMIT=m CONFIG_NETFILTER_XT_MATCH_MAC=m @@ -931,8 +933,6 @@ CONFIG_NETFILTER_XT_MATCH_RECENT=m CONFIG_NETFILTER_XT_MATCH_SCTP=m CONFIG_NETFILTER_XT_MATCH_SOCKET=m CONFIG_NETFILTER_XT_MATCH_STATE=m -CONFIG_NETFILTER_XT_MATCH_LAYER7=m -# CONFIG_NETFILTER_XT_MATCH_LAYER7_DEBUG is not set CONFIG_NETFILTER_XT_MATCH_STATISTIC=m CONFIG_NETFILTER_XT_MATCH_STRING=m CONFIG_NETFILTER_XT_MATCH_TCPMSS=m @@ -1140,6 +1140,7 @@ CONFIG_NET_SCH_CHOKE=m CONFIG_NET_SCH_QFQ=m CONFIG_NET_SCH_CODEL=m CONFIG_NET_SCH_FQ_CODEL=m +CONFIG_NET_SCH_PIE=m CONFIG_NET_SCH_INGRESS=m CONFIG_NET_SCH_PLUG=m diff --git a/config/kernel/kernel.config.i586-ipfire-pae b/config/kernel/kernel.config.i586-ipfire-pae index 8dc185bca..88e557af9 100644 --- a/config/kernel/kernel.config.i586-ipfire-pae +++ b/config/kernel/kernel.config.i586-ipfire-pae @@ -1152,6 +1152,7 @@ CONFIG_NET_SCH_CHOKE=m CONFIG_NET_SCH_QFQ=m CONFIG_NET_SCH_CODEL=m CONFIG_NET_SCH_FQ_CODEL=m +CONFIG_NET_SCH_PIE=m CONFIG_NET_SCH_INGRESS=m CONFIG_NET_SCH_PLUG=m diff --git a/config/rootfiles/common/armv5tel/linux-kirkwood b/config/rootfiles/common/armv5tel/linux-kirkwood index 9cb3057d0..4880b60ff 100644 --- a/config/rootfiles/common/armv5tel/linux-kirkwood +++ b/config/rootfiles/common/armv5tel/linux-kirkwood @@ -2243,6 +2243,7 @@ lib/modules/KVER-ipfire-kirkwood #lib/modules/KVER-ipfire-kirkwood/kernel/net/sched/sch_mqprio.ko #lib/modules/KVER-ipfire-kirkwood/kernel/net/sched/sch_multiq.ko #lib/modules/KVER-ipfire-kirkwood/kernel/net/sched/sch_netem.ko +#lib/modules/KVER-ipfire-kirkwood/kernel/net/sched/sch_pie.ko #lib/modules/KVER-ipfire-kirkwood/kernel/net/sched/sch_plug.ko #lib/modules/KVER-ipfire-kirkwood/kernel/net/sched/sch_prio.ko #lib/modules/KVER-ipfire-kirkwood/kernel/net/sched/sch_qfq.ko diff --git a/config/rootfiles/common/armv5tel/linux-multi b/config/rootfiles/common/armv5tel/linux-multi index 1fd840c95..3c20c5236 100644 --- a/config/rootfiles/common/armv5tel/linux-multi +++ b/config/rootfiles/common/armv5tel/linux-multi @@ -2195,6 +2195,7 @@ lib/modules/KVER-ipfire-multi #lib/modules/KVER-ipfire-multi/kernel/net/sched/sch_mqprio.ko #lib/modules/KVER-ipfire-multi/kernel/net/sched/sch_multiq.ko #lib/modules/KVER-ipfire-multi/kernel/net/sched/sch_netem.ko +#lib/modules/KVER-ipfire-multi/kernel/net/sched/sch_pie.ko #lib/modules/KVER-ipfire-multi/kernel/net/sched/sch_plug.ko #lib/modules/KVER-ipfire-multi/kernel/net/sched/sch_prio.ko #lib/modules/KVER-ipfire-multi/kernel/net/sched/sch_qfq.ko diff --git a/config/rootfiles/common/armv5tel/linux-rpi b/config/rootfiles/common/armv5tel/linux-rpi index 2ea9efe19..01f9312ee 100644 --- a/config/rootfiles/common/armv5tel/linux-rpi +++ b/config/rootfiles/common/armv5tel/linux-rpi @@ -1804,6 +1804,7 @@ lib/modules/KVER-ipfire-rpi #lib/modules/KVER-ipfire-rpi/kernel/net/sched/sch_mqprio.ko #lib/modules/KVER-ipfire-rpi/kernel/net/sched/sch_multiq.ko #lib/modules/KVER-ipfire-rpi/kernel/net/sched/sch_netem.ko +#lib/modules/KVER-ipfire-rpi/kernel/net/sched/sch_pie.ko #lib/modules/KVER-ipfire-rpi/kernel/net/sched/sch_plug.ko #lib/modules/KVER-ipfire-rpi/kernel/net/sched/sch_prio.ko #lib/modules/KVER-ipfire-rpi/kernel/net/sched/sch_qfq.ko diff --git a/config/rootfiles/common/i586/linux b/config/rootfiles/common/i586/linux index 48bd9ced5..fac263157 100644 --- a/config/rootfiles/common/i586/linux +++ b/config/rootfiles/common/i586/linux @@ -2663,6 +2663,7 @@ lib/modules/KVER-ipfire #lib/modules/KVER-ipfire/kernel/net/sched/sch_mqprio.ko #lib/modules/KVER-ipfire/kernel/net/sched/sch_multiq.ko #lib/modules/KVER-ipfire/kernel/net/sched/sch_netem.ko +#lib/modules/KVER-ipfire/kernel/net/sched/sch_pie.ko #lib/modules/KVER-ipfire/kernel/net/sched/sch_plug.ko #lib/modules/KVER-ipfire/kernel/net/sched/sch_prio.ko #lib/modules/KVER-ipfire/kernel/net/sched/sch_qfq.ko diff --git a/config/rootfiles/packages/linux-pae b/config/rootfiles/packages/linux-pae index eaaa3e589..fdba462f2 100644 --- a/config/rootfiles/packages/linux-pae +++ b/config/rootfiles/packages/linux-pae @@ -2675,6 +2675,7 @@ lib/modules/KVER-ipfire-pae #lib/modules/KVER-ipfire-pae/kernel/net/sched/sch_mqprio.ko #lib/modules/KVER-ipfire-pae/kernel/net/sched/sch_multiq.ko #lib/modules/KVER-ipfire-pae/kernel/net/sched/sch_netem.ko +#lib/modules/KVER-ipfire-pae/kernel/net/sched/sch_pie.ko #lib/modules/KVER-ipfire-pae/kernel/net/sched/sch_plug.ko #lib/modules/KVER-ipfire-pae/kernel/net/sched/sch_prio.ko #lib/modules/KVER-ipfire-pae/kernel/net/sched/sch_qfq.ko diff --git a/lfs/linux b/lfs/linux index 2bf32d39b..497fd03b8 100644 --- a/lfs/linux +++ b/lfs/linux @@ -120,6 +120,9 @@ $(TARGET) : $(patsubst %,$(DIR_DL)/%,$(objects)) # Layer7-patch cd $(DIR_APP) && patch -Np1 < $(DIR_SRC)/src/patches/linux-3.10-layer7-filter.patch + # pie packet scheduler + cd $(DIR_APP) && patch -Np1 < $(DIR_SRC)/src/patches/linux-3.10.39-pie_packet_sch.patch + # Grsecurity-patches ifneq "$(KCFG)" "-headers" cd $(DIR_APP) && xz -c -d $(DIR_DL)/$(GRS_PATCHES) | patch -Np1 diff --git a/src/patches/linux-3.10.39-pie_packet_sch.patch b/src/patches/linux-3.10.39-pie_packet_sch.patch new file mode 100644 index 000000000..6bec37450 --- /dev/null +++ b/src/patches/linux-3.10.39-pie_packet_sch.patch @@ -0,0 +1,674 @@ +From d4b36210c2e6ecef0ce52fb6c18c51144f5c2d88 Mon Sep 17 00:00:00 2001 +From: Vijay Subramanian +Date: Sat, 4 Jan 2014 17:33:55 -0800 +Subject: net: pkt_sched: PIE AQM scheme + +Proportional Integral controller Enhanced (PIE) is a scheduler to address the +bufferbloat problem. + +>From the IETF draft below: +" Bufferbloat is a phenomenon where excess buffers in the network cause high +latency and jitter. As more and more interactive applications (e.g. voice over +IP, real time video streaming and financial transactions) run in the Internet, +high latency and jitter degrade application performance. There is a pressing +need to design intelligent queue management schemes that can control latency and +jitter; and hence provide desirable quality of service to users. + +We present here a lightweight design, PIE(Proportional Integral controller +Enhanced) that can effectively control the average queueing latency to a target +value. Simulation results, theoretical analysis and Linux testbed results have +shown that PIE can ensure low latency and achieve high link utilization under +various congestion situations. The design does not require per-packet +timestamp, so it incurs very small overhead and is simple enough to implement +in both hardware and software. " + +Many thanks to Dave Taht for extensive feedback, reviews, testing and +suggestions. Thanks also to Stephen Hemminger and Eric Dumazet for reviews and +suggestions. Naeem Khademi and Dave Taht independently contributed to ECN +support. + +For more information, please see technical paper about PIE in the IEEE +Conference on High Performance Switching and Routing 2013. A copy of the paper +can be found at ftp://ftpeng.cisco.com/pie/. + +Please also refer to the IETF draft submission at +http://tools.ietf.org/html/draft-pan-tsvwg-pie-00 + +All relevant code, documents and test scripts and results can be found at +ftp://ftpeng.cisco.com/pie/. + +For problems with the iproute2/tc or Linux kernel code, please contact Vijay +Subramanian (vijaynsu@cisco.com or subramanian.vijay@gmail.com) Mythili Prabhu +(mysuryan@cisco.com) + +Signed-off-by: Vijay Subramanian +Signed-off-by: Mythili Prabhu +CC: Dave Taht +Signed-off-by: David S. Miller + +diff -Naur linux-3.10.39.org/include/uapi/linux/pkt_sched.h linux-3.10.39/include/uapi/linux/pkt_sched.h +--- linux-3.10.39.org/include/uapi/linux/pkt_sched.h 2014-05-06 16:56:24.000000000 +0200 ++++ linux-3.10.39/include/uapi/linux/pkt_sched.h 2014-05-15 10:33:08.296828477 +0200 +@@ -744,4 +744,29 @@ + }; + }; + ++/* PIE */ ++enum { ++ TCA_PIE_UNSPEC, ++ TCA_PIE_TARGET, ++ TCA_PIE_LIMIT, ++ TCA_PIE_TUPDATE, ++ TCA_PIE_ALPHA, ++ TCA_PIE_BETA, ++ TCA_PIE_ECN, ++ TCA_PIE_BYTEMODE, ++ __TCA_PIE_MAX ++}; ++#define TCA_PIE_MAX (__TCA_PIE_MAX - 1) ++ ++struct tc_pie_xstats { ++ __u32 prob; /* current probability */ ++ __u32 delay; /* current delay in ms */ ++ __u32 avg_dq_rate; /* current average dq_rate in bits/pie_time */ ++ __u32 packets_in; /* total number of packets enqueued */ ++ __u32 dropped; /* packets dropped due to pie_action */ ++ __u32 overlimit; /* dropped due to lack of space in queue */ ++ __u32 maxq; /* maximum queue size */ ++ __u32 ecn_mark; /* packets marked with ecn*/ ++}; ++ + #endif +diff -Naur linux-3.10.39.org/net/sched/Kconfig linux-3.10.39/net/sched/Kconfig +--- linux-3.10.39.org/net/sched/Kconfig 2014-05-06 16:56:24.000000000 +0200 ++++ linux-3.10.39/net/sched/Kconfig 2014-05-15 09:30:29.866632326 +0200 +@@ -272,6 +272,19 @@ + + If unsure, say N. + ++config NET_SCH_PIE ++ tristate "Proportional Integral controller Enhanced (PIE) scheduler" ++ help ++ Say Y here if you want to use the Proportional Integral controller ++ Enhanced scheduler packet scheduling algorithm. ++ For more information, please see ++ http://tools.ietf.org/html/draft-pan-tsvwg-pie-00 ++ ++ To compile this driver as a module, choose M here: the module ++ will be called sch_pie. ++ ++ If unsure, say N. ++ + config NET_SCH_INGRESS + tristate "Ingress Qdisc" + depends on NET_CLS_ACT +diff -Naur linux-3.10.39.org/net/sched/Makefile linux-3.10.39/net/sched/Makefile +--- linux-3.10.39.org/net/sched/Makefile 2014-05-06 16:56:24.000000000 +0200 ++++ linux-3.10.39/net/sched/Makefile 2014-05-15 10:34:55.533502406 +0200 +@@ -39,6 +39,7 @@ + obj-$(CONFIG_NET_SCH_QFQ) += sch_qfq.o + obj-$(CONFIG_NET_SCH_CODEL) += sch_codel.o + obj-$(CONFIG_NET_SCH_FQ_CODEL) += sch_fq_codel.o ++obj-$(CONFIG_NET_SCH_PIE) += sch_pie.o + + obj-$(CONFIG_NET_CLS_U32) += cls_u32.o + obj-$(CONFIG_NET_CLS_ROUTE4) += cls_route.o +diff -Naur linux-3.10.39.org/net/sched/sch_pie.c linux-3.10.39/net/sched/sch_pie.c +--- linux-3.10.39.org/net/sched/sch_pie.c 1970-01-01 01:00:00.000000000 +0100 ++++ linux-3.10.39/net/sched/sch_pie.c 2014-05-15 09:30:29.869966724 +0200 +@@ -0,0 +1,555 @@ ++/* Copyright (C) 2013 Cisco Systems, Inc, 2013. ++ * ++ * This program is free software; you can redistribute it and/or ++ * modify it under the terms of the GNU General Public License ++ * as published by the Free Software Foundation; either version 2 ++ * of the License. ++ * ++ * This program is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * Author: Vijay Subramanian ++ * Author: Mythili Prabhu ++ * ++ * ECN support is added by Naeem Khademi ++ * University of Oslo, Norway. ++ */ ++ ++#include ++#include ++#include ++#include ++#include ++#include ++#include ++#include ++ ++#define QUEUE_THRESHOLD 10000 ++#define DQCOUNT_INVALID -1 ++#define MAX_PROB 0xffffffff ++#define PIE_SCALE 8 ++ ++/* parameters used */ ++struct pie_params { ++ psched_time_t target; /* user specified target delay in pschedtime */ ++ u32 tupdate; /* timer frequency (in jiffies) */ ++ u32 limit; /* number of packets that can be enqueued */ ++ u32 alpha; /* alpha and beta are between -4 and 4 */ ++ u32 beta; /* and are used for shift relative to 1 */ ++ bool ecn; /* true if ecn is enabled */ ++ bool bytemode; /* to scale drop early prob based on pkt size */ ++}; ++ ++/* variables used */ ++struct pie_vars { ++ u32 prob; /* probability but scaled by u32 limit. */ ++ psched_time_t burst_time; ++ psched_time_t qdelay; ++ psched_time_t qdelay_old; ++ u64 dq_count; /* measured in bytes */ ++ psched_time_t dq_tstamp; /* drain rate */ ++ u32 avg_dq_rate; /* bytes per pschedtime tick,scaled */ ++ u32 qlen_old; /* in bytes */ ++}; ++ ++/* statistics gathering */ ++struct pie_stats { ++ u32 packets_in; /* total number of packets enqueued */ ++ u32 dropped; /* packets dropped due to pie_action */ ++ u32 overlimit; /* dropped due to lack of space in queue */ ++ u32 maxq; /* maximum queue size */ ++ u32 ecn_mark; /* packets marked with ECN */ ++}; ++ ++/* private data for the Qdisc */ ++struct pie_sched_data { ++ struct pie_params params; ++ struct pie_vars vars; ++ struct pie_stats stats; ++ struct timer_list adapt_timer; ++}; ++ ++static void pie_params_init(struct pie_params *params) ++{ ++ params->alpha = 2; ++ params->beta = 20; ++ params->tupdate = usecs_to_jiffies(30 * USEC_PER_MSEC); /* 30 ms */ ++ params->limit = 1000; /* default of 1000 packets */ ++ params->target = PSCHED_NS2TICKS(20 * NSEC_PER_MSEC); /* 20 ms */ ++ params->ecn = false; ++ params->bytemode = false; ++} ++ ++static void pie_vars_init(struct pie_vars *vars) ++{ ++ vars->dq_count = DQCOUNT_INVALID; ++ vars->avg_dq_rate = 0; ++ /* default of 100 ms in pschedtime */ ++ vars->burst_time = PSCHED_NS2TICKS(100 * NSEC_PER_MSEC); ++} ++ ++static bool drop_early(struct Qdisc *sch, u32 packet_size) ++{ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ u32 rnd; ++ u32 local_prob = q->vars.prob; ++ u32 mtu = psched_mtu(qdisc_dev(sch)); ++ ++ /* If there is still burst allowance left skip random early drop */ ++ if (q->vars.burst_time > 0) ++ return false; ++ ++ /* If current delay is less than half of target, and ++ * if drop prob is low already, disable early_drop ++ */ ++ if ((q->vars.qdelay < q->params.target / 2) ++ && (q->vars.prob < MAX_PROB / 5)) ++ return false; ++ ++ /* If we have fewer than 2 mtu-sized packets, disable drop_early, ++ * similar to min_th in RED ++ */ ++ if (sch->qstats.backlog < 2 * mtu) ++ return false; ++ ++ /* If bytemode is turned on, use packet size to compute new ++ * probablity. Smaller packets will have lower drop prob in this case ++ */ ++ if (q->params.bytemode && packet_size <= mtu) ++ local_prob = (local_prob / mtu) * packet_size; ++ else ++ local_prob = q->vars.prob; ++ ++ rnd = net_random(); ++ if (rnd < local_prob) ++ return true; ++ ++ return false; ++} ++ ++static int pie_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch) ++{ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ bool enqueue = false; ++ ++ if (unlikely(qdisc_qlen(sch) >= sch->limit)) { ++ q->stats.overlimit++; ++ goto out; ++ } ++ ++ if (!drop_early(sch, skb->len)) { ++ enqueue = true; ++ } else if (q->params.ecn && (q->vars.prob <= MAX_PROB / 10) && ++ INET_ECN_set_ce(skb)) { ++ /* If packet is ecn capable, mark it if drop probability ++ * is lower than 10%, else drop it. ++ */ ++ q->stats.ecn_mark++; ++ enqueue = true; ++ } ++ ++ /* we can enqueue the packet */ ++ if (enqueue) { ++ q->stats.packets_in++; ++ if (qdisc_qlen(sch) > q->stats.maxq) ++ q->stats.maxq = qdisc_qlen(sch); ++ ++ return qdisc_enqueue_tail(skb, sch); ++ } ++ ++out: ++ q->stats.dropped++; ++ return qdisc_drop(skb, sch); ++} ++ ++static const struct nla_policy pie_policy[TCA_PIE_MAX + 1] = { ++ [TCA_PIE_TARGET] = {.type = NLA_U32}, ++ [TCA_PIE_LIMIT] = {.type = NLA_U32}, ++ [TCA_PIE_TUPDATE] = {.type = NLA_U32}, ++ [TCA_PIE_ALPHA] = {.type = NLA_U32}, ++ [TCA_PIE_BETA] = {.type = NLA_U32}, ++ [TCA_PIE_ECN] = {.type = NLA_U32}, ++ [TCA_PIE_BYTEMODE] = {.type = NLA_U32}, ++}; ++ ++static int pie_change(struct Qdisc *sch, struct nlattr *opt) ++{ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ struct nlattr *tb[TCA_PIE_MAX + 1]; ++ unsigned int qlen; ++ int err; ++ ++ if (!opt) ++ return -EINVAL; ++ ++ err = nla_parse_nested(tb, TCA_PIE_MAX, opt, pie_policy); ++ if (err < 0) ++ return err; ++ ++ sch_tree_lock(sch); ++ ++ /* convert from microseconds to pschedtime */ ++ if (tb[TCA_PIE_TARGET]) { ++ /* target is in us */ ++ u32 target = nla_get_u32(tb[TCA_PIE_TARGET]); ++ ++ /* convert to pschedtime */ ++ q->params.target = PSCHED_NS2TICKS((u64)target * NSEC_PER_USEC); ++ } ++ ++ /* tupdate is in jiffies */ ++ if (tb[TCA_PIE_TUPDATE]) ++ q->params.tupdate = usecs_to_jiffies(nla_get_u32(tb[TCA_PIE_TUPDATE])); ++ ++ if (tb[TCA_PIE_LIMIT]) { ++ u32 limit = nla_get_u32(tb[TCA_PIE_LIMIT]); ++ ++ q->params.limit = limit; ++ sch->limit = limit; ++ } ++ ++ if (tb[TCA_PIE_ALPHA]) ++ q->params.alpha = nla_get_u32(tb[TCA_PIE_ALPHA]); ++ ++ if (tb[TCA_PIE_BETA]) ++ q->params.beta = nla_get_u32(tb[TCA_PIE_BETA]); ++ ++ if (tb[TCA_PIE_ECN]) ++ q->params.ecn = nla_get_u32(tb[TCA_PIE_ECN]); ++ ++ if (tb[TCA_PIE_BYTEMODE]) ++ q->params.bytemode = nla_get_u32(tb[TCA_PIE_BYTEMODE]); ++ ++ /* Drop excess packets if new limit is lower */ ++ qlen = sch->q.qlen; ++ while (sch->q.qlen > sch->limit) { ++ struct sk_buff *skb = __skb_dequeue(&sch->q); ++ ++ sch->qstats.backlog -= qdisc_pkt_len(skb); ++ qdisc_drop(skb, sch); ++ } ++ qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen); ++ ++ sch_tree_unlock(sch); ++ return 0; ++} ++ ++static void pie_process_dequeue(struct Qdisc *sch, struct sk_buff *skb) ++{ ++ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ int qlen = sch->qstats.backlog; /* current queue size in bytes */ ++ ++ /* If current queue is about 10 packets or more and dq_count is unset ++ * we have enough packets to calculate the drain rate. Save ++ * current time as dq_tstamp and start measurement cycle. ++ */ ++ if (qlen >= QUEUE_THRESHOLD && q->vars.dq_count == DQCOUNT_INVALID) { ++ q->vars.dq_tstamp = psched_get_time(); ++ q->vars.dq_count = 0; ++ } ++ ++ /* Calculate the average drain rate from this value. If queue length ++ * has receded to a small value viz., <= QUEUE_THRESHOLD bytes,reset ++ * the dq_count to -1 as we don't have enough packets to calculate the ++ * drain rate anymore The following if block is entered only when we ++ * have a substantial queue built up (QUEUE_THRESHOLD bytes or more) ++ * and we calculate the drain rate for the threshold here. dq_count is ++ * in bytes, time difference in psched_time, hence rate is in ++ * bytes/psched_time. ++ */ ++ if (q->vars.dq_count != DQCOUNT_INVALID) { ++ q->vars.dq_count += skb->len; ++ ++ if (q->vars.dq_count >= QUEUE_THRESHOLD) { ++ psched_time_t now = psched_get_time(); ++ u32 dtime = now - q->vars.dq_tstamp; ++ u32 count = q->vars.dq_count << PIE_SCALE; ++ ++ if (dtime == 0) ++ return; ++ ++ count = count / dtime; ++ ++ if (q->vars.avg_dq_rate == 0) ++ q->vars.avg_dq_rate = count; ++ else ++ q->vars.avg_dq_rate = ++ (q->vars.avg_dq_rate - ++ (q->vars.avg_dq_rate >> 3)) + (count >> 3); ++ ++ /* If the queue has receded below the threshold, we hold ++ * on to the last drain rate calculated, else we reset ++ * dq_count to 0 to re-enter the if block when the next ++ * packet is dequeued ++ */ ++ if (qlen < QUEUE_THRESHOLD) ++ q->vars.dq_count = DQCOUNT_INVALID; ++ else { ++ q->vars.dq_count = 0; ++ q->vars.dq_tstamp = psched_get_time(); ++ } ++ ++ if (q->vars.burst_time > 0) { ++ if (q->vars.burst_time > dtime) ++ q->vars.burst_time -= dtime; ++ else ++ q->vars.burst_time = 0; ++ } ++ } ++ } ++} ++ ++static void calculate_probability(struct Qdisc *sch) ++{ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ u32 qlen = sch->qstats.backlog; /* queue size in bytes */ ++ psched_time_t qdelay = 0; /* in pschedtime */ ++ psched_time_t qdelay_old = q->vars.qdelay; /* in pschedtime */ ++ s32 delta = 0; /* determines the change in probability */ ++ u32 oldprob; ++ u32 alpha, beta; ++ bool update_prob = true; ++ ++ q->vars.qdelay_old = q->vars.qdelay; ++ ++ if (q->vars.avg_dq_rate > 0) ++ qdelay = (qlen << PIE_SCALE) / q->vars.avg_dq_rate; ++ else ++ qdelay = 0; ++ ++ /* If qdelay is zero and qlen is not, it means qlen is very small, less ++ * than dequeue_rate, so we do not update probabilty in this round ++ */ ++ if (qdelay == 0 && qlen != 0) ++ update_prob = false; ++ ++ /* Add ranges for alpha and beta, more aggressive for high dropping ++ * mode and gentle steps for light dropping mode ++ * In light dropping mode, take gentle steps; in medium dropping mode, ++ * take medium steps; in high dropping mode, take big steps. ++ */ ++ if (q->vars.prob < MAX_PROB / 100) { ++ alpha = ++ (q->params.alpha * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 7; ++ beta = ++ (q->params.beta * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 7; ++ } else if (q->vars.prob < MAX_PROB / 10) { ++ alpha = ++ (q->params.alpha * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 5; ++ beta = ++ (q->params.beta * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 5; ++ } else { ++ alpha = ++ (q->params.alpha * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 4; ++ beta = ++ (q->params.beta * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 4; ++ } ++ ++ /* alpha and beta should be between 0 and 32, in multiples of 1/16 */ ++ delta += alpha * ((qdelay - q->params.target)); ++ delta += beta * ((qdelay - qdelay_old)); ++ ++ oldprob = q->vars.prob; ++ ++ /* to ensure we increase probability in steps of no more than 2% */ ++ if (delta > (s32) (MAX_PROB / (100 / 2)) && ++ q->vars.prob >= MAX_PROB / 10) ++ delta = (MAX_PROB / 100) * 2; ++ ++ /* Non-linear drop: ++ * Tune drop probability to increase quickly for high delays(>= 250ms) ++ * 250ms is derived through experiments and provides error protection ++ */ ++ ++ if (qdelay > (PSCHED_NS2TICKS(250 * NSEC_PER_MSEC))) ++ delta += MAX_PROB / (100 / 2); ++ ++ q->vars.prob += delta; ++ ++ if (delta > 0) { ++ /* prevent overflow */ ++ if (q->vars.prob < oldprob) { ++ q->vars.prob = MAX_PROB; ++ /* Prevent normalization error. If probability is at ++ * maximum value already, we normalize it here, and ++ * skip the check to do a non-linear drop in the next ++ * section. ++ */ ++ update_prob = false; ++ } ++ } else { ++ /* prevent underflow */ ++ if (q->vars.prob > oldprob) ++ q->vars.prob = 0; ++ } ++ ++ /* Non-linear drop in probability: Reduce drop probability quickly if ++ * delay is 0 for 2 consecutive Tupdate periods. ++ */ ++ ++ if ((qdelay == 0) && (qdelay_old == 0) && update_prob) ++ q->vars.prob = (q->vars.prob * 98) / 100; ++ ++ q->vars.qdelay = qdelay; ++ q->vars.qlen_old = qlen; ++ ++ /* We restart the measurement cycle if the following conditions are met ++ * 1. If the delay has been low for 2 consecutive Tupdate periods ++ * 2. Calculated drop probability is zero ++ * 3. We have atleast one estimate for the avg_dq_rate ie., ++ * is a non-zero value ++ */ ++ if ((q->vars.qdelay < q->params.target / 2) && ++ (q->vars.qdelay_old < q->params.target / 2) && ++ (q->vars.prob == 0) && ++ (q->vars.avg_dq_rate > 0)) ++ pie_vars_init(&q->vars); ++} ++ ++static void pie_timer(unsigned long arg) ++{ ++ struct Qdisc *sch = (struct Qdisc *)arg; ++ struct pie_sched_data *q = qdisc_priv(sch); ++ spinlock_t *root_lock = qdisc_lock(qdisc_root_sleeping(sch)); ++ ++ spin_lock(root_lock); ++ calculate_probability(sch); ++ ++ /* reset the timer to fire after 'tupdate'. tupdate is in jiffies. */ ++ if (q->params.tupdate) ++ mod_timer(&q->adapt_timer, jiffies + q->params.tupdate); ++ spin_unlock(root_lock); ++ ++} ++ ++static int pie_init(struct Qdisc *sch, struct nlattr *opt) ++{ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ ++ pie_params_init(&q->params); ++ pie_vars_init(&q->vars); ++ sch->limit = q->params.limit; ++ ++ setup_timer(&q->adapt_timer, pie_timer, (unsigned long)sch); ++ mod_timer(&q->adapt_timer, jiffies + HZ / 2); ++ ++ if (opt) { ++ int err = pie_change(sch, opt); ++ ++ if (err) ++ return err; ++ } ++ ++ return 0; ++} ++ ++static int pie_dump(struct Qdisc *sch, struct sk_buff *skb) ++{ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ struct nlattr *opts; ++ ++ opts = nla_nest_start(skb, TCA_OPTIONS); ++ if (opts == NULL) ++ goto nla_put_failure; ++ ++ /* convert target from pschedtime to us */ ++ if (nla_put_u32(skb, TCA_PIE_TARGET, ++ ((u32) PSCHED_TICKS2NS(q->params.target)) / ++ NSEC_PER_USEC) || ++ nla_put_u32(skb, TCA_PIE_LIMIT, sch->limit) || ++ nla_put_u32(skb, TCA_PIE_TUPDATE, jiffies_to_usecs(q->params.tupdate)) || ++ nla_put_u32(skb, TCA_PIE_ALPHA, q->params.alpha) || ++ nla_put_u32(skb, TCA_PIE_BETA, q->params.beta) || ++ nla_put_u32(skb, TCA_PIE_ECN, q->params.ecn) || ++ nla_put_u32(skb, TCA_PIE_BYTEMODE, q->params.bytemode)) ++ goto nla_put_failure; ++ ++ return nla_nest_end(skb, opts); ++ ++nla_put_failure: ++ nla_nest_cancel(skb, opts); ++ return -1; ++ ++} ++ ++static int pie_dump_stats(struct Qdisc *sch, struct gnet_dump *d) ++{ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ struct tc_pie_xstats st = { ++ .prob = q->vars.prob, ++ .delay = ((u32) PSCHED_TICKS2NS(q->vars.qdelay)) / ++ NSEC_PER_USEC, ++ /* unscale and return dq_rate in bytes per sec */ ++ .avg_dq_rate = q->vars.avg_dq_rate * ++ (PSCHED_TICKS_PER_SEC) >> PIE_SCALE, ++ .packets_in = q->stats.packets_in, ++ .overlimit = q->stats.overlimit, ++ .maxq = q->stats.maxq, ++ .dropped = q->stats.dropped, ++ .ecn_mark = q->stats.ecn_mark, ++ }; ++ ++ return gnet_stats_copy_app(d, &st, sizeof(st)); ++} ++ ++static struct sk_buff *pie_qdisc_dequeue(struct Qdisc *sch) ++{ ++ struct sk_buff *skb; ++ skb = __qdisc_dequeue_head(sch, &sch->q); ++ ++ if (!skb) ++ return NULL; ++ ++ pie_process_dequeue(sch, skb); ++ return skb; ++} ++ ++static void pie_reset(struct Qdisc *sch) ++{ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ qdisc_reset_queue(sch); ++ pie_vars_init(&q->vars); ++} ++ ++static void pie_destroy(struct Qdisc *sch) ++{ ++ struct pie_sched_data *q = qdisc_priv(sch); ++ q->params.tupdate = 0; ++ del_timer_sync(&q->adapt_timer); ++} ++ ++static struct Qdisc_ops pie_qdisc_ops __read_mostly = { ++ .id = "pie", ++ .priv_size = sizeof(struct pie_sched_data), ++ .enqueue = pie_qdisc_enqueue, ++ .dequeue = pie_qdisc_dequeue, ++ .peek = qdisc_peek_dequeued, ++ .init = pie_init, ++ .destroy = pie_destroy, ++ .reset = pie_reset, ++ .change = pie_change, ++ .dump = pie_dump, ++ .dump_stats = pie_dump_stats, ++ .owner = THIS_MODULE, ++}; ++ ++static int __init pie_module_init(void) ++{ ++ return register_qdisc(&pie_qdisc_ops); ++} ++ ++static void __exit pie_module_exit(void) ++{ ++ unregister_qdisc(&pie_qdisc_ops); ++} ++ ++module_init(pie_module_init); ++module_exit(pie_module_exit); ++ ++MODULE_DESCRIPTION("Proportional Integral controller Enhanced (PIE) scheduler"); ++MODULE_AUTHOR("Vijay Subramanian"); ++MODULE_AUTHOR("Mythili Prabhu"); ++MODULE_LICENSE("GPL");