cregit-Linux how code gets into the kernel

Release 4.8 net/sched/sch_netem.c

Directory: net/sched
 * net/sched/sch_netem.c        Network emulator
 *              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.
 *              Many of the algorithms and ideas for this came from
 *              NIST Net which is not copyrighted.
 * Authors:     Stephen Hemminger <>
 *              Catalin(ux aka Dino) BOIE <catab at umbrella dot ro>

#include <linux/mm.h>
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/skbuff.h>
#include <linux/vmalloc.h>
#include <linux/rtnetlink.h>
#include <linux/reciprocal_div.h>
#include <linux/rbtree.h>

#include <net/netlink.h>
#include <net/pkt_sched.h>
#include <net/inet_ecn.h>

#define VERSION "1.3"

/*      Network Emulation Queuing algorithm.

        Sources: [1] Mark Carson, Darrin Santay, "NIST Net - A Linux-based
                 Network Emulation Tool
                 [2] Luigi Rizzo, DummyNet for FreeBSD


         This started out as a simple way to delay outgoing packets to
         test TCP but has grown to include most of the functionality
         of a full blown network emulator like NISTnet. It can delay
         packets and add random jitter (and correlation). The random
         distribution can be loaded from a table as well to provide
         normal, Pareto, or experimental curves. Packet loss,
         duplication, and reordering can also be emulated.

         This qdisc does not do classification that can be handled in
         layering other disciplines.  It does not need to do bandwidth
         control either since that can be handled by using token
         bucket or other rate control.

     Correlated Loss Generator models

        Added generation of correlated loss according to the
        "Gilbert-Elliot" model, a 4-state markov model.

        [1] NetemCLG Home
        [2] S. Salsano, F. Ludovici, A. Ordine, "Definition of a general
        and intuitive loss model for packet networks and its implementation
        in the Netem module in the Linux kernel", available in [1]

        Authors: Stefano Salsano <stefano.salsano at
                 Fabio Ludovici <fabio.ludovici at>

struct netem_sched_data {
	/* internal t(ime)fifo qdisc uses t_root and sch->limit */
struct rb_root t_root;

	/* optional qdisc for classful handling (NULL at netem init) */
struct Qdisc	*qdisc;

struct qdisc_watchdog watchdog;

psched_tdiff_t latency;
psched_tdiff_t jitter;

u32 loss;
u32 ecn;
u32 limit;
u32 counter;
u32 gap;
u32 duplicate;
u32 reorder;
u32 corrupt;
u64 rate;
s32 packet_overhead;
u32 cell_size;
struct reciprocal_value cell_size_reciprocal;
s32 cell_overhead;

struct crndstate {
u32 last;
u32 rho;

delay_cor, loss_cor, dup_cor, reorder_cor, corrupt_cor;

struct disttable {
u32  size;
s16 table[0];

	enum  {

	enum {

	enum {

	/* Correlated Loss Generation models */
struct clgstate {
		/* state of the Markov chain */
u8 state;

		/* 4-states and Gilbert-Elliot models */
u32 a1;	/* p13 for 4-states or p for GE */
u32 a2;	/* p31 for 4-states or r for GE */
u32 a3;	/* p32 for 4-states or h for GE */
u32 a4;	/* p14 for 4-states or 1-k for GE */
u32 a5; /* p23 used only in 4-states */


/* Time stamp put into socket buffer control block
 * Only valid when skbs are in our internal t(ime)fifo queue.
 * As skb->rbnode uses same storage than skb->next, skb->prev and skb->tstamp,
 * and skb->next & skb->prev are scratch space for a qdisc,
 * we save skb->tstamp value in skb->cb[] before destroying it.

struct netem_skb_cb {
psched_time_t	time_to_send;
ktime_t		tstamp_save;

static struct sk_buff *netem_rb_to_skb(struct rb_node *rb) { return container_of(rb, struct sk_buff, rbnode); }


eric dumazeteric dumazet24100.00%2100.00%

static inline struct netem_skb_cb *netem_skb_cb(struct sk_buff *skb) { /* we assume we can use skb next/prev/tstamp as storage for rb_node */ qdisc_cb_private_validate(skb, sizeof(struct netem_skb_cb)); return (struct netem_skb_cb *)qdisc_skb_cb(skb)->data; }


jussi kivilinnajussi kivilinna3692.31%250.00%
david s. millerdavid s. miller25.13%125.00%
eric dumazeteric dumazet12.56%125.00%

/* init_crandom - initialize correlated random number generator * Use entropy source for initial seed. */
static void init_crandom(struct crndstate *state, unsigned long rho) { state->rho = rho; state->last = prandom_u32(); }


stephen hemmingerstephen hemminger2796.43%150.00%

/* get_crandom - correlated random number generator * Next number depends on last value. * rho is scaled to avoid floating point. */
static u32 get_crandom(struct crndstate *state) { u64 value, rho; unsigned long answer; if (state->rho == 0) /* no correlation */ return prandom_u32(); value = prandom_u32(); rho = (u64)state->rho + 1; answer = (value * ((1ull<<32) - rho) + state->last * rho) >> 32; state->last = answer; return answer; }


stephen hemmingerstephen hemminger8097.56%480.00%

/* loss_4state - 4-state model loss generator * Generates losses according to the 4-state Markov chain adopted in * the GI (General and Intuitive) loss model. */
static bool loss_4state(struct netem_sched_data *q) { struct clgstate *clg = &q->clg; u32 rnd = prandom_u32(); /* * Makes a comparison between rnd and the transition * probabilities outgoing from the current state, then decides the * next state and if the next packet has to be transmitted or lost. * The four states correspond to: * TX_IN_GAP_PERIOD => successfully transmitted packets within a gap period * LOST_IN_BURST_PERIOD => isolated losses within a gap period * LOST_IN_GAP_PERIOD => lost packets within a burst period * TX_IN_GAP_PERIOD => successfully transmitted packets within a burst period */ switch (clg->state) { case TX_IN_GAP_PERIOD: if (rnd < clg->a4) { clg->state = LOST_IN_BURST_PERIOD; return true; } else if (clg->a4 < rnd && rnd < clg->a1 + clg->a4) { clg->state = LOST_IN_GAP_PERIOD; return true; } else if (clg->a1 + clg->a4 < rnd) { clg->state = TX_IN_GAP_PERIOD; } break; case TX_IN_BURST_PERIOD: if (rnd < clg->a5) { clg->state = LOST_IN_GAP_PERIOD; return true; } else { clg->state = TX_IN_BURST_PERIOD; } break; case LOST_IN_GAP_PERIOD: if (rnd < clg->a3) clg->state = TX_IN_BURST_PERIOD; else if (clg->a3 < rnd && rnd < clg->a2 + clg->a3) { clg->state = TX_IN_GAP_PERIOD; } else if (clg->a2 + clg->a3 < rnd) { clg->state = LOST_IN_GAP_PERIOD; return true; } break; case LOST_IN_BURST_PERIOD: clg->state = TX_IN_GAP_PERIOD; break; } return false; }


stephen hemmingerstephen hemminger20591.52%250.00%
yang yingliangyang yingliang188.04%125.00%

/* loss_gilb_ell - Gilbert-Elliot model loss generator * Generates losses according to the Gilbert-Elliot loss model or * its special cases (Gilbert or Simple Gilbert) * * Makes a comparison between random number and the transition * probabilities outgoing from the current state, then decides the * next state. A second random number is extracted and the comparison * with the loss probability of the current state decides if the next * packet will be transmitted or lost. */
static bool loss_gilb_ell(struct netem_sched_data *q) { struct clgstate *clg = &q->clg; switch (clg->state) { case GOOD_STATE: if (prandom_u32() < clg->a1) clg->state = BAD_STATE; if (prandom_u32() < clg->a4) return true; break; case BAD_STATE: if (prandom_u32() < clg->a2) clg->state = GOOD_STATE; if (prandom_u32() > clg->a3) return true; } return false; }


stephen hemmingerstephen hemminger8591.40%360.00%
yang yingliangyang yingliang44.30%120.00%

static bool loss_event(struct netem_sched_data *q) { switch (q->loss_model) { case CLG_RANDOM: /* Random packet drop 0 => none, ~0 => all */ return q->loss && q->loss >= get_crandom(&q->loss_cor); case CLG_4_STATES: /* 4state loss model algorithm (used also for GI model) * Extracts a value from the markov 4 state loss generator, * if it is 1 drops a packet and if needed writes the event in * the kernel logs */ return loss_4state(q); case CLG_GILB_ELL: /* Gilbert-Elliot loss model algorithm * Extracts a value from the Gilbert-Elliot loss generator, * if it is 1 drops a packet and if needed writes the event in * the kernel logs */ return loss_gilb_ell(q); } return false; /* not reached */ }


stephen hemmingerstephen hemminger64100.00%1100.00%

/* tabledist - return a pseudo-randomly distributed value with mean mu and * std deviation sigma. Uses table lookup to approximate the desired * distribution, and a uniformly-distributed pseudo-random source. */
static psched_tdiff_t tabledist(psched_tdiff_t mu, psched_tdiff_t sigma, struct crndstate *state, const struct disttable *dist) { psched_tdiff_t x; long t; u32 rnd; if (sigma == 0) return mu; rnd = get_crandom(state); /* default uniform distribution */ if (dist == NULL) return (rnd % (2*sigma)) - sigma + mu; t = dist->table[rnd % dist->size]; x = (sigma % NETEM_DIST_SCALE) * t; if (x >= 0) x += NETEM_DIST_SCALE/2; else x -= NETEM_DIST_SCALE/2; return x / NETEM_DIST_SCALE + (sigma / NETEM_DIST_SCALE) * t + mu; }


stephen hemmingerstephen hemminger127100.00%5100.00%

static psched_time_t packet_len_2_sched_time(unsigned int len, struct netem_sched_data *q) { u64 ticks; len += q->packet_overhead; if (q->cell_size) { u32 cells = reciprocal_divide(len, q->cell_size_reciprocal); if (len > cells * q->cell_size) /* extra cell needed for remainder */ cells++; len = cells * (q->cell_size + q->cell_overhead); } ticks = (u64)len * NSEC_PER_SEC; do_div(ticks, q->rate); return PSCHED_NS2TICKS(ticks); }


hagen paul pfeiferhagen paul pfeifer8386.46%266.67%
eric dumazeteric dumazet1313.54%133.33%

static void tfifo_reset(struct Qdisc *sch) { struct netem_sched_data *q = qdisc_priv(sch); struct rb_node *p; while ((p = rb_first(&q->t_root))) { struct sk_buff *skb = netem_rb_to_skb(p); rb_erase(p, &q->t_root); rtnl_kfree_skbs(skb, skb); } }


stephen hemmingerstephen hemminger6695.65%150.00%
eric dumazeteric dumazet34.35%150.00%

static void tfifo_enqueue(struct sk_buff *nskb, struct Qdisc *sch) { struct netem_sched_data *q = qdisc_priv(sch); psched_time_t tnext = netem_skb_cb(nskb)->time_to_send; struct rb_node **p = &q->t_root.rb_node, *parent = NULL; while (*p) { struct sk_buff *skb; parent = *p; skb = netem_rb_to_skb(parent); if (tnext >= netem_skb_cb(skb)->time_to_send) p = &parent->rb_right; else p = &parent->rb_left; } rb_link_node(&nskb->rbnode, parent, p); rb_insert_color(&nskb->rbnode, &q->t_root); sch->q.qlen++; }


eric dumazeteric dumazet136100.00%4100.00%

/* netem can't properly corrupt a megapacket (like we get from GSO), so instead * when we statistically choose to corrupt one, we instead segment it, returning * the first packet to be corrupted, and re-enqueue the remaining frames */
static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) { struct sk_buff *segs; netdev_features_t features = netif_skb_features(skb); segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); if (IS_ERR_OR_NULL(segs)) { qdisc_drop(skb, sch, to_free); return NULL; } consume_skb(skb); return segs; }


neil hormanneil horman6988.46%133.33%
eric dumazeteric dumazet810.26%133.33%
florian westphalflorian westphal11.28%133.33%

/* * Insert one skb into qdisc. * Note: parent depends on return value to account for queue length. * NET_XMIT_DROP: queue length didn't change. * NET_XMIT_SUCCESS: one skb was queued. */
static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) { struct netem_sched_data *q = qdisc_priv(sch); /* We don't fill cb now as skb_unshare() may invalidate it */ struct netem_skb_cb *cb; struct sk_buff *skb2; struct sk_buff *segs = NULL; unsigned int len = 0, last_len, prev_len = qdisc_pkt_len(skb); int nb = 0; int count = 1; int rc = NET_XMIT_SUCCESS; /* Random duplication */ if (q->duplicate && q->duplicate >= get_crandom(&q->dup_cor)) ++count; /* Drop packet? */ if (loss_event(q)) { if (q->ecn && INET_ECN_set_ce(skb)) qdisc_qstats_drop(sch); /* mark packet */ else --count; } if (count == 0) { qdisc_qstats_drop(sch); __qdisc_drop(skb, to_free); return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; } /* If a delay is expected, orphan the skb. (orphaning usually takes * place at TX completion time, so _before_ the link transit delay) */ if (q->latency || q->jitter) skb_orphan_partial(skb); /* * If we need to duplicate packet, then re-insert at top of the * qdisc tree, since parent queuer expects that only one * skb will be queued. */ if (count > 1 && (skb2 = skb_clone(skb, GFP_ATOMIC)) != NULL) { struct Qdisc *rootq = qdisc_root(sch); u32 dupsave = q->duplicate; /* prevent duplicating a dup... */ q->duplicate = 0; rootq->enqueue(skb2, rootq, to_free); q->duplicate = dupsave; } /* * Randomized packet corruption. * Make copy if needed since we are modifying * If packet is going to be hardware checksummed, then * do it now in software before we mangle it. */ if (q->corrupt && q->corrupt >= get_crandom(&q->corrupt_cor)) { if (skb_is_gso(skb)) { segs = netem_segment(skb, sch, to_free); if (!segs) return NET_XMIT_DROP; } else { segs = skb; } skb = segs; segs = segs->next; skb = skb_unshare(skb, GFP_ATOMIC); if (unlikely(!skb)) { qdisc_qstats_drop(sch); goto finish_segs; } if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb)) { qdisc_drop(skb, sch, to_free); goto finish_segs; } skb->data[prandom_u32() % skb_headlen(skb)] ^= 1<<(prandom_u32() % 8); } if (unlikely(skb_queue_len(&sch->q) >= sch->limit)) return qdisc_drop(skb, sch, to_free); qdisc_qstats_backlog_inc(sch, skb); cb = netem_skb_cb(skb); if (q->gap == 0 || /* not doing reordering */ q->counter < q->gap - 1 || /* inside last reordering gap */ q->reorder < get_crandom(&q->reorder_cor)) { psched_time_t now; psched_tdiff_t delay; delay = tabledist(q->latency, q->jitter, &q->delay_cor, q->delay_dist); now = psched_get_time(); if (q->rate) { struct sk_buff *last; if (!skb_queue_empty(&sch->q)) last = skb_peek_tail(&sch->q); else last = netem_rb_to_skb(rb_last(&q->t_root)); if (last) { /* * Last packet in queue is reference point (now), * calculate this time bonus and subtract * from delay. */ delay -= netem_skb_cb(last)->time_to_send - now; delay = max_t(psched_tdiff_t, 0, delay); now = netem_skb_cb(last)->time_to_send; } delay += packet_len_2_sched_time(qdisc_pkt_len(skb), q); } cb->time_to_send = now + delay; cb->tstamp_save = skb->tstamp; ++q->counter; tfifo_enqueue(skb, sch); } else { /* * Do re-ordering by putting one out of N packets at the front * of the queue. */ cb->time_to_send = psched_get_time(); q->counter = 0; __skb_queue_head(&sch->q, skb); sch->qstats.requeues++; } finish_segs: if (segs) { while (segs) { skb2 = segs->next; segs->next = NULL; qdisc_skb_cb(segs)->pkt_len = segs->len; last_len = segs->len; rc = qdisc_enqueue(segs, sch, to_free); if (rc != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(rc)) qdisc_qstats_drop(sch); } else { nb++; len += last_len; } segs = skb2; } sch->q.qlen += nb; if (nb > 1) qdisc_tree_reduce_backlog(sch, 1 - nb, prev_len - len); } return NET_XMIT_SUCCESS; }


stephen hemmingerstephen hemminger29838.95%1736.17%
neil hormanneil horman19225.10%12.13%
eric dumazeteric dumazet15320.00%1021.28%
hagen paul pfeiferhagen paul pfeifer374.84%24.26%
johannes naabjohannes naab233.01%12.13%
jarek poplawskijarek poplawski162.09%24.26%
patrick mchardypatrick mchardy111.44%36.38%
john fastabendjohn fastabend91.18%12.13%
david s. millerdavid s. miller70.92%24.26%
guillaume chazarainguillaume chazarain60.78%12.13%
yang yingliangyang yingliang30.39%12.13%
joe perchesjoe perches20.26%12.13%
jussi kivilinnajussi kivilinna20.26%12.13%
vijay subramanianvijay subramanian20.26%12.13%
florian westphalflorian westphal10.13%12.13%
hideaki yoshifujihideaki yoshifuji10.13%12.13%

static struct sk_buff *netem_dequeue(struct Qdisc *sch) { struct netem_sched_data *q = qdisc_priv(sch); struct sk_buff *skb; struct rb_node *p; tfifo_dequeue: skb = __skb_dequeue(&sch->q); if (skb) { qdisc_qstats_backlog_dec(sch, skb); deliver: qdisc_bstats_update(sch, skb); return skb; } p = rb_first(&q->t_root); if (p) { psched_time_t time_to_send; skb = netem_rb_to_skb(p); /* if more time remaining? */ time_to_send = netem_skb_cb(skb)->time_to_send; if (time_to_send <= psched_get_time()) { rb_erase(p, &q->t_root); sch->q.qlen--; qdisc_qstats_backlog_dec(sch, skb); skb->next = NULL; skb->prev = NULL; skb->tstamp = netem_skb_cb(skb)->tstamp_save; #ifdef CONFIG_NET_CLS_ACT /* * If it's at ingress let's pretend the delay is * from the network (tstamp will be updated). */ if (G_TC_FROM(skb->tc_verd) & AT_INGRESS) skb->tstamp.tv64 = 0; #endif if (q->qdisc) { unsigned int pkt_len = qdisc_pkt_len(skb); struct sk_buff *to_free = NULL; int err; err = qdisc_enqueue(skb, q->qdisc, &to_free); kfree_skb_list(to_free); if (err != NET_XMIT_SUCCESS && net_xmit_drop_count(err)) { qdisc_qstats_drop(sch); qdisc_tree_reduce_backlog(sch, 1, pkt_len); } goto tfifo_dequeue; } goto deliver; } if (q->qdisc) { skb = q->qdisc->ops->dequeue(q->qdisc); if (skb) goto deliver; } qdisc_watchdog_schedule(&q->watchdog, time_to_send); } if (q->qdisc) { skb = q->qdisc->ops->dequeue(q->qdisc); if (skb) goto deliver; } return NULL; }


eric dumazeteric dumazet22867.26%728.00%
stephen hemmingerstephen hemminger6218.29%1040.00%
jarek poplawskijarek poplawski298.55%312.00%
joseph beshayjoseph beshay92.65%14.00%
john fastabendjohn fastabend61.77%14.00%
americo wangamerico wang20.59%14.00%
jussi kivilinnajussi kivilinna20.59%14.00%
patrick mchardypatrick mchardy10.29%14.00%

static void netem_reset(struct Qdisc *sch) { struct netem_sched_data *q = qdisc_priv(sch); qdisc_reset_queue(sch); tfifo_reset(sch); if (q->qdisc) qdisc_reset(q->qdisc); qdisc_watchdog_cancel(&q->watchdog); }


stephen hemmingerstephen hemminger3975.00%571.43%
eric dumazeteric dumazet1121.15%114.29%
patrick mchardypatrick mchardy23.85%114.29%

static void dist_free(struct disttable *d) { kvfree(d); }


stephen hemmingerstephen hemminger1593.75%150.00%
americo wangamerico wang16.25%150.00%

/* * Distribution data is a variable size payload containing * signed 16 bit values. */
static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr) { struct netem_sched_data *q = qdisc_priv(sch); size_t n = nla_len(attr)/sizeof(__s16); const __s16 *data = nla_data(attr); spinlock_t *root_lock; struct disttable *d; int i; size_t s; if (n > NETEM_DIST_MAX) return -EINVAL; s = sizeof(struct disttable) + n * sizeof(s16); d = kmalloc(s, GFP_KERNEL | __GFP_NOWARN); if (!d) d = vmalloc(s); if (!d) return -ENOMEM; d->size = n; for (i = 0; i < n; i++) d->table[i] = data[i]; root_lock = qdisc_root_sleeping_lock(sch); spin_lock_bh(root_lock); swap(q->delay_dist, d); spin_unlock_bh(root_lock); dist_free(d); return 0; }


stephen hemmingerstephen hemminger15985.03%861.54%
david s. millerdavid s. miller126.42%17.69%
eric dumazeteric dumazet105.35%17.69%
patrick mchardypatrick mchardy52.67%215.38%
jarek poplawskijarek poplawski10.53%17.69%

static void get_correlation(struct netem_sched_data *q, const struct nlattr *attr) { const struct tc_netem_corr *c = nla_data(attr); init_crandom(&q->delay_cor, c->delay_corr); init_crandom(&q->loss_cor, c->loss_corr); init_crandom(&q->dup_cor, c->dup_corr); }


stephen hemmingerstephen hemminger6093.75%466.67%
yang yingliangyang yingliang23.12%116.67%
patrick mchardypatrick mchardy23.12%116.67%

static void get_reorder(struct netem_sched_data *q, const struct nlattr *attr) { const struct tc_netem_reorder *r = nla_data(attr); q->reorder = r->probability; init_crandom(&q->reorder_cor, r->correlation); }


stephen hemmingerstephen hemminger4491.67%250.00%
patrick mchardypatrick mchardy24.17%125.00%
yang yingliangyang yingliang24.17%125.00%

static void get_corrupt(struct netem_sched_data *q, const struct nlattr *attr) { const struct tc_netem_corrupt *r = nla_data(attr); q->corrupt = r->probability; init_crandom(&q->corrupt_cor, r->correlation); }


stephen hemmingerstephen hemminger4491.67%360.00%
patrick mchardypatrick mchardy24.17%120.00%
yang yingliangyang yingliang24.17%120.00%

static void get_rate(struct netem_sched_data *q, const struct nlattr *attr) { const struct tc_netem_rate *r = nla_data(attr); q->rate = r->rate; q->packet_overhead = r->packet_overhead; q->cell_size = r->cell_size;