From 13d2a1d2b032de08d7dcab6a1edcd47802681f96 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 20 Nov 2008 04:10:00 -0800 Subject: [PATCH] pkt_sched: add DRR scheduler Add classful DRR scheduler as a more flexible replacement for SFQ. The main difference to the algorithm described in "Efficient Fair Queueing using Deficit Round Robin" is that this implementation doesn't drop packets from the longest queue on overrun because its classful and limits are handled by each individual child qdisc. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/pkt_sched.h | 16 ++ net/sched/Kconfig | 11 + net/sched/Makefile | 1 + net/sched/sch_drr.c | 505 ++++++++++++++++++++++++++++++++++++++ 4 files changed, 533 insertions(+) create mode 100644 net/sched/sch_drr.c diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h index 5d921fa91a5..e3f133adba7 100644 --- a/include/linux/pkt_sched.h +++ b/include/linux/pkt_sched.h @@ -500,4 +500,20 @@ struct tc_netem_corrupt #define NETEM_DIST_SCALE 8192 +/* DRR */ + +enum +{ + TCA_DRR_UNSPEC, + TCA_DRR_QUANTUM, + __TCA_DRR_MAX +}; + +#define TCA_DRR_MAX (__TCA_DRR_MAX - 1) + +struct tc_drr_stats +{ + u32 deficit; +}; + #endif diff --git a/net/sched/Kconfig b/net/sched/Kconfig index 36543b6fcef..4f7ef0db302 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -194,6 +194,17 @@ config NET_SCH_NETEM If unsure, say N. +config NET_SCH_DRR + tristate "Deficit Round Robin scheduler (DRR)" + help + Say Y here if you want to use the Deficit Round Robin (DRR) packet + scheduling algorithm. + + To compile this driver as a module, choose M here: the module + will be called sch_drr. + + If unsure, say N. + config NET_SCH_INGRESS tristate "Ingress Qdisc" depends on NET_CLS_ACT diff --git a/net/sched/Makefile b/net/sched/Makefile index 70b35f8708c..54d950cd4b8 100644 --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -30,6 +30,7 @@ obj-$(CONFIG_NET_SCH_PRIO) += sch_prio.o obj-$(CONFIG_NET_SCH_MULTIQ) += sch_multiq.o obj-$(CONFIG_NET_SCH_ATM) += sch_atm.o obj-$(CONFIG_NET_SCH_NETEM) += sch_netem.o +obj-$(CONFIG_NET_SCH_DRR) += sch_drr.o obj-$(CONFIG_NET_CLS_U32) += cls_u32.o obj-$(CONFIG_NET_CLS_ROUTE4) += cls_route.o obj-$(CONFIG_NET_CLS_FW) += cls_fw.o diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c new file mode 100644 index 00000000000..e71a5de23e2 --- /dev/null +++ b/net/sched/sch_drr.c @@ -0,0 +1,505 @@ +/* + * net/sched/sch_drr.c Deficit Round Robin scheduler + * + * Copyright (c) 2008 Patrick McHardy + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * version 2 as published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +struct drr_class { + struct Qdisc_class_common common; + unsigned int refcnt; + unsigned int filter_cnt; + + struct gnet_stats_basic bstats; + struct gnet_stats_queue qstats; + struct gnet_stats_rate_est rate_est; + struct list_head alist; + struct Qdisc *qdisc; + + u32 quantum; + u32 deficit; +}; + +struct drr_sched { + struct list_head active; + struct tcf_proto *filter_list; + struct Qdisc_class_hash clhash; +}; + +static struct drr_class *drr_find_class(struct Qdisc *sch, u32 classid) +{ + struct drr_sched *q = qdisc_priv(sch); + struct Qdisc_class_common *clc; + + clc = qdisc_class_find(&q->clhash, classid); + if (clc == NULL) + return NULL; + return container_of(clc, struct drr_class, common); +} + +static void drr_purge_queue(struct drr_class *cl) +{ + unsigned int len = cl->qdisc->q.qlen; + + qdisc_reset(cl->qdisc); + qdisc_tree_decrease_qlen(cl->qdisc, len); +} + +static const struct nla_policy drr_policy[TCA_DRR_MAX + 1] = { + [TCA_DRR_QUANTUM] = { .type = NLA_U32 }, +}; + +static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid, + struct nlattr **tca, unsigned long *arg) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl = (struct drr_class *)*arg; + struct nlattr *tb[TCA_DRR_MAX + 1]; + u32 quantum; + int err; + + err = nla_parse_nested(tb, TCA_DRR_MAX, tca[TCA_OPTIONS], drr_policy); + if (err < 0) + return err; + + if (tb[TCA_DRR_QUANTUM]) { + quantum = nla_get_u32(tb[TCA_DRR_QUANTUM]); + if (quantum == 0) + return -EINVAL; + } else + quantum = psched_mtu(qdisc_dev(sch)); + + if (cl != NULL) { + sch_tree_lock(sch); + if (tb[TCA_DRR_QUANTUM]) + cl->quantum = quantum; + sch_tree_unlock(sch); + + if (tca[TCA_RATE]) + gen_replace_estimator(&cl->bstats, &cl->rate_est, + qdisc_root_sleeping_lock(sch), + tca[TCA_RATE]); + return 0; + } + + cl = kzalloc(sizeof(struct drr_class), GFP_KERNEL); + if (cl == NULL) + return -ENOBUFS; + + cl->refcnt = 1; + cl->common.classid = classid; + cl->quantum = quantum; + cl->qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, + &pfifo_qdisc_ops, classid); + if (cl->qdisc == NULL) + cl->qdisc = &noop_qdisc; + + if (tca[TCA_RATE]) + gen_replace_estimator(&cl->bstats, &cl->rate_est, + qdisc_root_sleeping_lock(sch), + tca[TCA_RATE]); + + sch_tree_lock(sch); + qdisc_class_hash_insert(&q->clhash, &cl->common); + sch_tree_unlock(sch); + + qdisc_class_hash_grow(sch, &q->clhash); + + *arg = (unsigned long)cl; + return 0; +} + +static void drr_destroy_class(struct Qdisc *sch, struct drr_class *cl) +{ + gen_kill_estimator(&cl->bstats, &cl->rate_est); + qdisc_destroy(cl->qdisc); + kfree(cl); +} + +static int drr_delete_class(struct Qdisc *sch, unsigned long arg) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl = (struct drr_class *)arg; + + if (cl->filter_cnt > 0) + return -EBUSY; + + sch_tree_lock(sch); + + drr_purge_queue(cl); + qdisc_class_hash_remove(&q->clhash, &cl->common); + + if (--cl->refcnt == 0) + drr_destroy_class(sch, cl); + + sch_tree_unlock(sch); + return 0; +} + +static unsigned long drr_get_class(struct Qdisc *sch, u32 classid) +{ + struct drr_class *cl = drr_find_class(sch, classid); + + if (cl != NULL) + cl->refcnt++; + + return (unsigned long)cl; +} + +static void drr_put_class(struct Qdisc *sch, unsigned long arg) +{ + struct drr_class *cl = (struct drr_class *)arg; + + if (--cl->refcnt == 0) + drr_destroy_class(sch, cl); +} + +static struct tcf_proto **drr_tcf_chain(struct Qdisc *sch, unsigned long cl) +{ + struct drr_sched *q = qdisc_priv(sch); + + if (cl) + return NULL; + + return &q->filter_list; +} + +static unsigned long drr_bind_tcf(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + struct drr_class *cl = drr_find_class(sch, classid); + + if (cl != NULL) + cl->filter_cnt++; + + return (unsigned long)cl; +} + +static void drr_unbind_tcf(struct Qdisc *sch, unsigned long arg) +{ + struct drr_class *cl = (struct drr_class *)arg; + + cl->filter_cnt--; +} + +static int drr_graft_class(struct Qdisc *sch, unsigned long arg, + struct Qdisc *new, struct Qdisc **old) +{ + struct drr_class *cl = (struct drr_class *)arg; + + if (new == NULL) { + new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue, + &pfifo_qdisc_ops, cl->common.classid); + if (new == NULL) + new = &noop_qdisc; + } + + sch_tree_lock(sch); + drr_purge_queue(cl); + *old = xchg(&cl->qdisc, new); + sch_tree_unlock(sch); + return 0; +} + +static struct Qdisc *drr_class_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct drr_class *cl = (struct drr_class *)arg; + + return cl->qdisc; +} + +static void drr_qlen_notify(struct Qdisc *csh, unsigned long arg) +{ + struct drr_class *cl = (struct drr_class *)arg; + + if (cl->qdisc->q.qlen == 0) + list_del(&cl->alist); +} + +static int drr_dump_class(struct Qdisc *sch, unsigned long arg, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct drr_class *cl = (struct drr_class *)arg; + struct nlattr *nest; + + tcm->tcm_parent = TC_H_ROOT; + tcm->tcm_handle = cl->common.classid; + tcm->tcm_info = cl->qdisc->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + NLA_PUT_U32(skb, TCA_DRR_QUANTUM, cl->quantum); + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + +static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg, + struct gnet_dump *d) +{ + struct drr_class *cl = (struct drr_class *)arg; + struct tc_drr_stats xstats; + + memset(&xstats, 0, sizeof(xstats)); + if (cl->qdisc->q.qlen) + xstats.deficit = cl->deficit; + + if (gnet_stats_copy_basic(d, &cl->bstats) < 0 || + gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 || + gnet_stats_copy_queue(d, &cl->qdisc->qstats) < 0) + return -1; + + return gnet_stats_copy_app(d, &xstats, sizeof(xstats)); +} + +static void drr_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + struct hlist_node *n; + unsigned int i; + + if (arg->stop) + return; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, n, &q->clhash.hash[i], common.hnode) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, (unsigned long)cl, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } +} + +static struct drr_class *drr_classify(struct sk_buff *skb, struct Qdisc *sch, + int *qerr) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + struct tcf_result res; + int result; + + if (TC_H_MAJ(skb->priority ^ sch->handle) == 0) { + cl = drr_find_class(sch, skb->priority); + if (cl != NULL) + return cl; + } + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + result = tc_classify(skb, q->filter_list, &res); + if (result >= 0) { +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_QUEUED: + case TC_ACT_STOLEN: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + case TC_ACT_SHOT: + return NULL; + } +#endif + cl = (struct drr_class *)res.class; + if (cl == NULL) + cl = drr_find_class(sch, res.classid); + return cl; + } + return NULL; +} + +static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + unsigned int len; + int err; + + cl = drr_classify(skb, sch, &err); + if (cl == NULL) { + if (err & __NET_XMIT_BYPASS) + sch->qstats.drops++; + kfree_skb(skb); + return err; + } + + len = qdisc_pkt_len(skb); + err = qdisc_enqueue(skb, cl->qdisc); + if (unlikely(err != NET_XMIT_SUCCESS)) { + if (net_xmit_drop_count(err)) { + cl->qstats.drops++; + sch->qstats.drops++; + } + return err; + } + + if (cl->qdisc->q.qlen == 1) { + list_add_tail(&cl->alist, &q->active); + cl->deficit = cl->quantum; + } + + cl->bstats.packets++; + cl->bstats.bytes += len; + sch->bstats.packets++; + sch->bstats.bytes += len; + + sch->q.qlen++; + return err; +} + +static struct sk_buff *drr_dequeue(struct Qdisc *sch) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + struct sk_buff *skb; + unsigned int len; + + while (!list_empty(&q->active)) { + cl = list_first_entry(&q->active, struct drr_class, alist); + skb = cl->qdisc->ops->peek(cl->qdisc); + if (skb == NULL) + goto skip; + + len = qdisc_pkt_len(skb); + if (len <= cl->deficit) { + cl->deficit -= len; + skb = qdisc_dequeue_peeked(cl->qdisc); + if (cl->qdisc->q.qlen == 0) + list_del(&cl->alist); + sch->q.qlen--; + return skb; + } + + cl->deficit += cl->quantum; +skip: + list_move_tail(&cl->alist, &q->active); + } + return NULL; +} + +static unsigned int drr_drop(struct Qdisc *sch) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + unsigned int len; + + list_for_each_entry(cl, &q->active, alist) { + if (cl->qdisc->ops->drop) { + len = cl->qdisc->ops->drop(cl->qdisc); + if (len > 0) { + if (cl->qdisc->q.qlen == 0) + list_del(&cl->alist); + return len; + } + } + } + return 0; +} + +static int drr_init_qdisc(struct Qdisc *sch, struct nlattr *opt) +{ + struct drr_sched *q = qdisc_priv(sch); + int err; + + err = qdisc_class_hash_init(&q->clhash); + if (err < 0) + return err; + INIT_LIST_HEAD(&q->active); + return 0; +} + +static void drr_reset_qdisc(struct Qdisc *sch) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + struct hlist_node *n; + unsigned int i; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, n, &q->clhash.hash[i], common.hnode) { + if (cl->qdisc->q.qlen) + list_del(&cl->alist); + qdisc_reset(cl->qdisc); + } + } + sch->q.qlen = 0; +} + +static void drr_destroy_qdisc(struct Qdisc *sch) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + struct hlist_node *n, *next; + unsigned int i; + + tcf_destroy_chain(&q->filter_list); + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry_safe(cl, n, next, &q->clhash.hash[i], + common.hnode) + drr_destroy_class(sch, cl); + } + qdisc_class_hash_destroy(&q->clhash); +} + +static const struct Qdisc_class_ops drr_class_ops = { + .change = drr_change_class, + .delete = drr_delete_class, + .get = drr_get_class, + .put = drr_put_class, + .tcf_chain = drr_tcf_chain, + .bind_tcf = drr_bind_tcf, + .unbind_tcf = drr_unbind_tcf, + .graft = drr_graft_class, + .leaf = drr_class_leaf, + .qlen_notify = drr_qlen_notify, + .dump = drr_dump_class, + .dump_stats = drr_dump_class_stats, + .walk = drr_walk, +}; + +static struct Qdisc_ops drr_qdisc_ops __read_mostly = { + .cl_ops = &drr_class_ops, + .id = "drr", + .priv_size = sizeof(struct drr_sched), + .enqueue = drr_enqueue, + .dequeue = drr_dequeue, + .peek = qdisc_peek_dequeued, + .drop = drr_drop, + .init = drr_init_qdisc, + .reset = drr_reset_qdisc, + .destroy = drr_destroy_qdisc, + .owner = THIS_MODULE, +}; + +static int __init drr_init(void) +{ + return register_qdisc(&drr_qdisc_ops); +} + +static void __exit drr_exit(void) +{ + unregister_qdisc(&drr_qdisc_ops); +} + +module_init(drr_init); +module_exit(drr_exit); +MODULE_LICENSE("GPL"); -- 2.41.1