sch_cbq.c

linux 内核源代码

rtattr_failure:
	nlmsg_trim(skb, b);
	return -1;
}

static int
cbq_dump_stats(struct Qdisc *sch, struct gnet_dump *d)
{
	struct cbq_sched_data *q = qdisc_priv(sch);

	q->link.xstats.avgidle = q->link.avgidle;
	return gnet_stats_copy_app(d, &q->link.xstats, sizeof(q->link.xstats));
}

static int
cbq_dump_class(struct Qdisc *sch, unsigned long arg,
	       struct sk_buff *skb, struct tcmsg *tcm)
{
	struct cbq_class *cl = (struct cbq_class*)arg;
	unsigned char *b = skb_tail_pointer(skb);
	struct rtattr *rta;

	if (cl->tparent)
		tcm->tcm_parent = cl->tparent->classid;
	else
		tcm->tcm_parent = TC_H_ROOT;
	tcm->tcm_handle = cl->classid;
	tcm->tcm_info = cl->q->handle;

	rta = (struct rtattr*)b;
	RTA_PUT(skb, TCA_OPTIONS, 0, NULL);
	if (cbq_dump_attr(skb, cl) < 0)
		goto rtattr_failure;
	rta->rta_len = skb_tail_pointer(skb) - b;
	return skb->len;

rtattr_failure:
	nlmsg_trim(skb, b);
	return -1;
}

static int
cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
	struct gnet_dump *d)
{
	struct cbq_sched_data *q = qdisc_priv(sch);
	struct cbq_class *cl = (struct cbq_class*)arg;

	cl->qstats.qlen = cl->q->q.qlen;
	cl->xstats.avgidle = cl->avgidle;
	cl->xstats.undertime = 0;

	if (cl->undertime != PSCHED_PASTPERFECT)
		cl->xstats.undertime = cl->undertime - q->now;

	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->qstats) < 0)
		return -1;

	return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
}

static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
		     struct Qdisc **old)
{
	struct cbq_class *cl = (struct cbq_class*)arg;

	if (cl) {
		if (new == NULL) {
			if ((new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
						     cl->classid)) == NULL)
				return -ENOBUFS;
		} else {
#ifdef CONFIG_NET_CLS_ACT
			if (cl->police == TC_POLICE_RECLASSIFY)
				new->reshape_fail = cbq_reshape_fail;
#endif
		}
		sch_tree_lock(sch);
		*old = xchg(&cl->q, new);
		qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
		qdisc_reset(*old);
		sch_tree_unlock(sch);

		return 0;
	}
	return -ENOENT;
}

static struct Qdisc *
cbq_leaf(struct Qdisc *sch, unsigned long arg)
{
	struct cbq_class *cl = (struct cbq_class*)arg;

	return cl ? cl->q : NULL;
}

static void cbq_qlen_notify(struct Qdisc *sch, unsigned long arg)
{
	struct cbq_class *cl = (struct cbq_class *)arg;

	if (cl->q->q.qlen == 0)
		cbq_deactivate_class(cl);
}

static unsigned long cbq_get(struct Qdisc *sch, u32 classid)
{
	struct cbq_sched_data *q = qdisc_priv(sch);
	struct cbq_class *cl = cbq_class_lookup(q, classid);

	if (cl) {
		cl->refcnt++;
		return (unsigned long)cl;
	}
	return 0;
}

static void cbq_destroy_class(struct Qdisc *sch, struct cbq_class *cl)
{
	struct cbq_sched_data *q = qdisc_priv(sch);

	BUG_TRAP(!cl->filters);

	tcf_destroy_chain(cl->filter_list);
	qdisc_destroy(cl->q);
	qdisc_put_rtab(cl->R_tab);
	gen_kill_estimator(&cl->bstats, &cl->rate_est);
	if (cl != &q->link)
		kfree(cl);
}

static void
cbq_destroy(struct Qdisc* sch)
{
	struct cbq_sched_data *q = qdisc_priv(sch);
	struct cbq_class *cl;
	unsigned h;

#ifdef CONFIG_NET_CLS_ACT
	q->rx_class = NULL;
#endif
	/*
	 * Filters must be destroyed first because we don't destroy the
	 * classes from root to leafs which means that filters can still
	 * be bound to classes which have been destroyed already. --TGR '04
	 */
	for (h = 0; h < 16; h++) {
		for (cl = q->classes[h]; cl; cl = cl->next) {
			tcf_destroy_chain(cl->filter_list);
			cl->filter_list = NULL;
		}
	}
	for (h = 0; h < 16; h++) {
		struct cbq_class *next;

		for (cl = q->classes[h]; cl; cl = next) {
			next = cl->next;
			cbq_destroy_class(sch, cl);
		}
	}
}

static void cbq_put(struct Qdisc *sch, unsigned long arg)
{
	struct cbq_class *cl = (struct cbq_class*)arg;

	if (--cl->refcnt == 0) {
#ifdef CONFIG_NET_CLS_ACT
		struct cbq_sched_data *q = qdisc_priv(sch);

		spin_lock_bh(&sch->dev->queue_lock);
		if (q->rx_class == cl)
			q->rx_class = NULL;
		spin_unlock_bh(&sch->dev->queue_lock);
#endif

		cbq_destroy_class(sch, cl);
	}
}

static int
cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct rtattr **tca,
		 unsigned long *arg)
{
	int err;
	struct cbq_sched_data *q = qdisc_priv(sch);
	struct cbq_class *cl = (struct cbq_class*)*arg;
	struct rtattr *opt = tca[TCA_OPTIONS-1];
	struct rtattr *tb[TCA_CBQ_MAX];
	struct cbq_class *parent;
	struct qdisc_rate_table *rtab = NULL;

	if (opt==NULL || rtattr_parse_nested(tb, TCA_CBQ_MAX, opt))
		return -EINVAL;

	if (tb[TCA_CBQ_OVL_STRATEGY-1] &&
	    RTA_PAYLOAD(tb[TCA_CBQ_OVL_STRATEGY-1]) < sizeof(struct tc_cbq_ovl))
		return -EINVAL;

	if (tb[TCA_CBQ_FOPT-1] &&
	    RTA_PAYLOAD(tb[TCA_CBQ_FOPT-1]) < sizeof(struct tc_cbq_fopt))
		return -EINVAL;

	if (tb[TCA_CBQ_RATE-1] &&
	    RTA_PAYLOAD(tb[TCA_CBQ_RATE-1]) < sizeof(struct tc_ratespec))
			return -EINVAL;

	if (tb[TCA_CBQ_LSSOPT-1] &&
	    RTA_PAYLOAD(tb[TCA_CBQ_LSSOPT-1]) < sizeof(struct tc_cbq_lssopt))
			return -EINVAL;

	if (tb[TCA_CBQ_WRROPT-1] &&
	    RTA_PAYLOAD(tb[TCA_CBQ_WRROPT-1]) < sizeof(struct tc_cbq_wrropt))
			return -EINVAL;

#ifdef CONFIG_NET_CLS_ACT
	if (tb[TCA_CBQ_POLICE-1] &&
	    RTA_PAYLOAD(tb[TCA_CBQ_POLICE-1]) < sizeof(struct tc_cbq_police))
			return -EINVAL;
#endif

	if (cl) {
		/* Check parent */
		if (parentid) {
			if (cl->tparent && cl->tparent->classid != parentid)
				return -EINVAL;
			if (!cl->tparent && parentid != TC_H_ROOT)
				return -EINVAL;
		}

		if (tb[TCA_CBQ_RATE-1]) {
			rtab = qdisc_get_rtab(RTA_DATA(tb[TCA_CBQ_RATE-1]), tb[TCA_CBQ_RTAB-1]);
			if (rtab == NULL)
				return -EINVAL;
		}

		/* Change class parameters */
		sch_tree_lock(sch);

		if (cl->next_alive != NULL)
			cbq_deactivate_class(cl);

		if (rtab) {
			rtab = xchg(&cl->R_tab, rtab);
			qdisc_put_rtab(rtab);
		}

		if (tb[TCA_CBQ_LSSOPT-1])
			cbq_set_lss(cl, RTA_DATA(tb[TCA_CBQ_LSSOPT-1]));

		if (tb[TCA_CBQ_WRROPT-1]) {
			cbq_rmprio(q, cl);
			cbq_set_wrr(cl, RTA_DATA(tb[TCA_CBQ_WRROPT-1]));
		}

		if (tb[TCA_CBQ_OVL_STRATEGY-1])
			cbq_set_overlimit(cl, RTA_DATA(tb[TCA_CBQ_OVL_STRATEGY-1]));

#ifdef CONFIG_NET_CLS_ACT
		if (tb[TCA_CBQ_POLICE-1])
			cbq_set_police(cl, RTA_DATA(tb[TCA_CBQ_POLICE-1]));
#endif

		if (tb[TCA_CBQ_FOPT-1])
			cbq_set_fopt(cl, RTA_DATA(tb[TCA_CBQ_FOPT-1]));

		if (cl->q->q.qlen)
			cbq_activate_class(cl);

		sch_tree_unlock(sch);

		if (tca[TCA_RATE-1])
			gen_replace_estimator(&cl->bstats, &cl->rate_est,
					      &sch->dev->queue_lock,
					      tca[TCA_RATE-1]);
		return 0;
	}

	if (parentid == TC_H_ROOT)
		return -EINVAL;

	if (tb[TCA_CBQ_WRROPT-1] == NULL || tb[TCA_CBQ_RATE-1] == NULL ||
	    tb[TCA_CBQ_LSSOPT-1] == NULL)
		return -EINVAL;

	rtab = qdisc_get_rtab(RTA_DATA(tb[TCA_CBQ_RATE-1]), tb[TCA_CBQ_RTAB-1]);
	if (rtab == NULL)
		return -EINVAL;

	if (classid) {
		err = -EINVAL;
		if (TC_H_MAJ(classid^sch->handle) || cbq_class_lookup(q, classid))
			goto failure;
	} else {
		int i;
		classid = TC_H_MAKE(sch->handle,0x8000);

		for (i=0; i<0x8000; i++) {
			if (++q->hgenerator >= 0x8000)
				q->hgenerator = 1;
			if (cbq_class_lookup(q, classid|q->hgenerator) == NULL)
				break;
		}
		err = -ENOSR;
		if (i >= 0x8000)
			goto failure;
		classid = classid|q->hgenerator;
	}

	parent = &q->link;
	if (parentid) {
		parent = cbq_class_lookup(q, parentid);
		err = -EINVAL;
		if (parent == NULL)
			goto failure;
	}

	err = -ENOBUFS;
	cl = kzalloc(sizeof(*cl), GFP_KERNEL);
	if (cl == NULL)
		goto failure;
	cl->R_tab = rtab;
	rtab = NULL;
	cl->refcnt = 1;
	if (!(cl->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid)))
		cl->q = &noop_qdisc;
	cl->classid = classid;
	cl->tparent = parent;
	cl->qdisc = sch;
	cl->allot = parent->allot;
	cl->quantum = cl->allot;
	cl->weight = cl->R_tab->rate.rate;

	sch_tree_lock(sch);
	cbq_link_class(cl);
	cl->borrow = cl->tparent;
	if (cl->tparent != &q->link)
		cl->share = cl->tparent;
	cbq_adjust_levels(parent);
	cl->minidle = -0x7FFFFFFF;
	cbq_set_lss(cl, RTA_DATA(tb[TCA_CBQ_LSSOPT-1]));
	cbq_set_wrr(cl, RTA_DATA(tb[TCA_CBQ_WRROPT-1]));
	if (cl->ewma_log==0)
		cl->ewma_log = q->link.ewma_log;
	if (cl->maxidle==0)
		cl->maxidle = q->link.maxidle;
	if (cl->avpkt==0)
		cl->avpkt = q->link.avpkt;
	cl->overlimit = cbq_ovl_classic;
	if (tb[TCA_CBQ_OVL_STRATEGY-1])
		cbq_set_overlimit(cl, RTA_DATA(tb[TCA_CBQ_OVL_STRATEGY-1]));
#ifdef CONFIG_NET_CLS_ACT
	if (tb[TCA_CBQ_POLICE-1])
		cbq_set_police(cl, RTA_DATA(tb[TCA_CBQ_POLICE-1]));
#endif
	if (tb[TCA_CBQ_FOPT-1])
		cbq_set_fopt(cl, RTA_DATA(tb[TCA_CBQ_FOPT-1]));
	sch_tree_unlock(sch);

	if (tca[TCA_RATE-1])
		gen_new_estimator(&cl->bstats, &cl->rate_est,
				  &sch->dev->queue_lock, tca[TCA_RATE-1]);

	*arg = (unsigned long)cl;
	return 0;

failure:
	qdisc_put_rtab(rtab);
	return err;
}

static int cbq_delete(struct Qdisc *sch, unsigned long arg)
{
	struct cbq_sched_data *q = qdisc_priv(sch);
	struct cbq_class *cl = (struct cbq_class*)arg;
	unsigned int qlen;

	if (cl->filters || cl->children || cl == &q->link)
		return -EBUSY;

	sch_tree_lock(sch);

	qlen = cl->q->q.qlen;
	qdisc_reset(cl->q);
	qdisc_tree_decrease_qlen(cl->q, qlen);

	if (cl->next_alive)
		cbq_deactivate_class(cl);

	if (q->tx_borrowed == cl)
		q->tx_borrowed = q->tx_class;
	if (q->tx_class == cl) {
		q->tx_class = NULL;
		q->tx_borrowed = NULL;
	}
#ifdef CONFIG_NET_CLS_ACT
	if (q->rx_class == cl)
		q->rx_class = NULL;
#endif

	cbq_unlink_class(cl);
	cbq_adjust_levels(cl->tparent);
	cl->defmap = 0;
	cbq_sync_defmap(cl);

	cbq_rmprio(q, cl);
	sch_tree_unlock(sch);

	if (--cl->refcnt == 0)
		cbq_destroy_class(sch, cl);

	return 0;
}

static struct tcf_proto **cbq_find_tcf(struct Qdisc *sch, unsigned long arg)
{
	struct cbq_sched_data *q = qdisc_priv(sch);
	struct cbq_class *cl = (struct cbq_class *)arg;

	if (cl == NULL)
		cl = &q->link;

	return &cl->filter_list;
}

static unsigned long cbq_bind_filter(struct Qdisc *sch, unsigned long parent,
				     u32 classid)
{
	struct cbq_sched_data *q = qdisc_priv(sch);
	struct cbq_class *p = (struct cbq_class*)parent;
	struct cbq_class *cl = cbq_class_lookup(q, classid);

	if (cl) {
		if (p && p->level <= cl->level)
			return 0;
		cl->filters++;
		return (unsigned long)cl;
	}
	return 0;
}

static void cbq_unbind_filter(struct Qdisc *sch, unsigned long arg)
{
	struct cbq_class *cl = (struct cbq_class*)arg;

	cl->filters--;
}

static void cbq_walk(struct Qdisc *sch, struct qdisc_walker *arg)
{
	struct cbq_sched_data *q = qdisc_priv(sch);
	unsigned h;

	if (arg->stop)
		return;

	for (h = 0; h < 16; h++) {
		struct cbq_class *cl;

		for (cl = q->classes[h]; cl; cl = cl->next) {
			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 Qdisc_class_ops cbq_class_ops = {
	.graft		=	cbq_graft,
	.leaf		=	cbq_leaf,
	.qlen_notify	=	cbq_qlen_notify,
	.get		=	cbq_get,
	.put		=	cbq_put,
	.change		=	cbq_change_class,
	.delete		=	cbq_delete,
	.walk		=	cbq_walk,
	.tcf_chain	=	cbq_find_tcf,
	.bind_tcf	=	cbq_bind_filter,
	.unbind_tcf	=	cbq_unbind_filter,
	.dump		=	cbq_dump_class,
	.dump_stats	=	cbq_dump_class_stats,
};

static struct Qdisc_ops cbq_qdisc_ops = {
	.next		=	NULL,
	.cl_ops		=	&cbq_class_ops,
	.id		=	"cbq",
	.priv_size	=	sizeof(struct cbq_sched_data),
	.enqueue	=	cbq_enqueue,
	.dequeue	=	cbq_dequeue,
	.requeue	=	cbq_requeue,
	.drop		=	cbq_drop,
	.init		=	cbq_init,
	.reset		=	cbq_reset,
	.destroy	=	cbq_destroy,
	.change		=	NULL,
	.dump		=	cbq_dump,
	.dump_stats	=	cbq_dump_stats,
	.owner		=	THIS_MODULE,
};

static int __init cbq_module_init(void)
{
	return register_qdisc(&cbq_qdisc_ops);
}
static void __exit cbq_module_exit(void)
{
	unregister_qdisc(&cbq_qdisc_ops);
}
module_init(cbq_module_init)
module_exit(cbq_module_exit)
MODULE_LICENSE("GPL");