sch_api.c

linux 内核源代码

		if (!clid)
			return -EINVAL;
		if (q->handle == 0)
			return -ENOENT;
		if ((err = qdisc_graft(dev, p, clid, NULL, &q)) != 0)
			return err;
		if (q) {
			qdisc_notify(skb, n, clid, q, NULL);
			qdisc_lock_tree(dev);
			qdisc_destroy(q);
			qdisc_unlock_tree(dev);
		}
	} else {
		qdisc_notify(skb, n, clid, NULL, q);
	}
	return 0;
}

/*
   Create/change qdisc.
 */

static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
{
	struct tcmsg *tcm;
	struct rtattr **tca;
	struct net_device *dev;
	u32 clid;
	struct Qdisc *q, *p;
	int err;

replay:
	/* Reinit, just in case something touches this. */
	tcm = NLMSG_DATA(n);
	tca = arg;
	clid = tcm->tcm_parent;
	q = p = NULL;

	if ((dev = __dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
		return -ENODEV;

	if (clid) {
		if (clid != TC_H_ROOT) {
			if (clid != TC_H_INGRESS) {
				if ((p = qdisc_lookup(dev, TC_H_MAJ(clid))) == NULL)
					return -ENOENT;
				q = qdisc_leaf(p, clid);
			} else { /*ingress */
				q = dev->qdisc_ingress;
			}
		} else {
			q = dev->qdisc_sleeping;
		}

		/* It may be default qdisc, ignore it */
		if (q && q->handle == 0)
			q = NULL;

		if (!q || !tcm->tcm_handle || q->handle != tcm->tcm_handle) {
			if (tcm->tcm_handle) {
				if (q && !(n->nlmsg_flags&NLM_F_REPLACE))
					return -EEXIST;
				if (TC_H_MIN(tcm->tcm_handle))
					return -EINVAL;
				if ((q = qdisc_lookup(dev, tcm->tcm_handle)) == NULL)
					goto create_n_graft;
				if (n->nlmsg_flags&NLM_F_EXCL)
					return -EEXIST;
				if (tca[TCA_KIND-1] && rtattr_strcmp(tca[TCA_KIND-1], q->ops->id))
					return -EINVAL;
				if (q == p ||
				    (p && check_loop(q, p, 0)))
					return -ELOOP;
				atomic_inc(&q->refcnt);
				goto graft;
			} else {
				if (q == NULL)
					goto create_n_graft;

				/* This magic test requires explanation.
				 *
				 *   We know, that some child q is already
				 *   attached to this parent and have choice:
				 *   either to change it or to create/graft new one.
				 *
				 *   1. We are allowed to create/graft only
				 *   if CREATE and REPLACE flags are set.
				 *
				 *   2. If EXCL is set, requestor wanted to say,
				 *   that qdisc tcm_handle is not expected
				 *   to exist, so that we choose create/graft too.
				 *
				 *   3. The last case is when no flags are set.
				 *   Alas, it is sort of hole in API, we
				 *   cannot decide what to do unambiguously.
				 *   For now we select create/graft, if
				 *   user gave KIND, which does not match existing.
				 */
				if ((n->nlmsg_flags&NLM_F_CREATE) &&
				    (n->nlmsg_flags&NLM_F_REPLACE) &&
				    ((n->nlmsg_flags&NLM_F_EXCL) ||
				     (tca[TCA_KIND-1] &&
				      rtattr_strcmp(tca[TCA_KIND-1], q->ops->id))))
					goto create_n_graft;
			}
		}
	} else {
		if (!tcm->tcm_handle)
			return -EINVAL;
		q = qdisc_lookup(dev, tcm->tcm_handle);
	}

	/* Change qdisc parameters */
	if (q == NULL)
		return -ENOENT;
	if (n->nlmsg_flags&NLM_F_EXCL)
		return -EEXIST;
	if (tca[TCA_KIND-1] && rtattr_strcmp(tca[TCA_KIND-1], q->ops->id))
		return -EINVAL;
	err = qdisc_change(q, tca);
	if (err == 0)
		qdisc_notify(skb, n, clid, NULL, q);
	return err;

create_n_graft:
	if (!(n->nlmsg_flags&NLM_F_CREATE))
		return -ENOENT;
	if (clid == TC_H_INGRESS)
		q = qdisc_create(dev, tcm->tcm_parent, tcm->tcm_parent,
				 tca, &err);
	else
		q = qdisc_create(dev, tcm->tcm_parent, tcm->tcm_handle,
				 tca, &err);
	if (q == NULL) {
		if (err == -EAGAIN)
			goto replay;
		return err;
	}

graft:
	if (1) {
		struct Qdisc *old_q = NULL;
		err = qdisc_graft(dev, p, clid, q, &old_q);
		if (err) {
			if (q) {
				qdisc_lock_tree(dev);
				qdisc_destroy(q);
				qdisc_unlock_tree(dev);
			}
			return err;
		}
		qdisc_notify(skb, n, clid, old_q, q);
		if (old_q) {
			qdisc_lock_tree(dev);
			qdisc_destroy(old_q);
			qdisc_unlock_tree(dev);
		}
	}
	return 0;
}

static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
			 u32 pid, u32 seq, u16 flags, int event)
{
	struct tcmsg *tcm;
	struct nlmsghdr  *nlh;
	unsigned char *b = skb_tail_pointer(skb);
	struct gnet_dump d;

	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*tcm), flags);
	tcm = NLMSG_DATA(nlh);
	tcm->tcm_family = AF_UNSPEC;
	tcm->tcm__pad1 = 0;
	tcm->tcm__pad2 = 0;
	tcm->tcm_ifindex = q->dev->ifindex;
	tcm->tcm_parent = clid;
	tcm->tcm_handle = q->handle;
	tcm->tcm_info = atomic_read(&q->refcnt);
	RTA_PUT(skb, TCA_KIND, IFNAMSIZ, q->ops->id);
	if (q->ops->dump && q->ops->dump(q, skb) < 0)
		goto rtattr_failure;
	q->qstats.qlen = q->q.qlen;

	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
			TCA_XSTATS, q->stats_lock, &d) < 0)
		goto rtattr_failure;

	if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0)
		goto rtattr_failure;

	if (gnet_stats_copy_basic(&d, &q->bstats) < 0 ||
	    gnet_stats_copy_rate_est(&d, &q->rate_est) < 0 ||
	    gnet_stats_copy_queue(&d, &q->qstats) < 0)
		goto rtattr_failure;

	if (gnet_stats_finish_copy(&d) < 0)
		goto rtattr_failure;

	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
	return skb->len;

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

static int qdisc_notify(struct sk_buff *oskb, struct nlmsghdr *n,
			u32 clid, struct Qdisc *old, struct Qdisc *new)
{
	struct sk_buff *skb;
	u32 pid = oskb ? NETLINK_CB(oskb).pid : 0;

	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
	if (!skb)
		return -ENOBUFS;

	if (old && old->handle) {
		if (tc_fill_qdisc(skb, old, clid, pid, n->nlmsg_seq, 0, RTM_DELQDISC) < 0)
			goto err_out;
	}
	if (new) {
		if (tc_fill_qdisc(skb, new, clid, pid, n->nlmsg_seq, old ? NLM_F_REPLACE : 0, RTM_NEWQDISC) < 0)
			goto err_out;
	}

	if (skb->len)
		return rtnetlink_send(skb, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);

err_out:
	kfree_skb(skb);
	return -EINVAL;
}

static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
{
	int idx, q_idx;
	int s_idx, s_q_idx;
	struct net_device *dev;
	struct Qdisc *q;

	s_idx = cb->args[0];
	s_q_idx = q_idx = cb->args[1];
	read_lock(&dev_base_lock);
	idx = 0;
	for_each_netdev(&init_net, dev) {
		if (idx < s_idx)
			goto cont;
		if (idx > s_idx)
			s_q_idx = 0;
		q_idx = 0;
		list_for_each_entry(q, &dev->qdisc_list, list) {
			if (q_idx < s_q_idx) {
				q_idx++;
				continue;
			}
			if (tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
					  cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
				goto done;
			q_idx++;
		}
cont:
		idx++;
	}

done:
	read_unlock(&dev_base_lock);

	cb->args[0] = idx;
	cb->args[1] = q_idx;

	return skb->len;
}



/************************************************
 *	Traffic classes manipulation.		*
 ************************************************/



static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
{
	struct tcmsg *tcm = NLMSG_DATA(n);
	struct rtattr **tca = arg;
	struct net_device *dev;
	struct Qdisc *q = NULL;
	struct Qdisc_class_ops *cops;
	unsigned long cl = 0;
	unsigned long new_cl;
	u32 pid = tcm->tcm_parent;
	u32 clid = tcm->tcm_handle;
	u32 qid = TC_H_MAJ(clid);
	int err;

	if ((dev = __dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
		return -ENODEV;

	/*
	   parent == TC_H_UNSPEC - unspecified parent.
	   parent == TC_H_ROOT   - class is root, which has no parent.
	   parent == X:0	 - parent is root class.
	   parent == X:Y	 - parent is a node in hierarchy.
	   parent == 0:Y	 - parent is X:Y, where X:0 is qdisc.

	   handle == 0:0	 - generate handle from kernel pool.
	   handle == 0:Y	 - class is X:Y, where X:0 is qdisc.
	   handle == X:Y	 - clear.
	   handle == X:0	 - root class.
	 */

	/* Step 1. Determine qdisc handle X:0 */

	if (pid != TC_H_ROOT) {
		u32 qid1 = TC_H_MAJ(pid);

		if (qid && qid1) {
			/* If both majors are known, they must be identical. */
			if (qid != qid1)
				return -EINVAL;
		} else if (qid1) {
			qid = qid1;
		} else if (qid == 0)
			qid = dev->qdisc_sleeping->handle;

		/* Now qid is genuine qdisc handle consistent
		   both with parent and child.

		   TC_H_MAJ(pid) still may be unspecified, complete it now.
		 */
		if (pid)
			pid = TC_H_MAKE(qid, pid);
	} else {
		if (qid == 0)
			qid = dev->qdisc_sleeping->handle;
	}

	/* OK. Locate qdisc */
	if ((q = qdisc_lookup(dev, qid)) == NULL)
		return -ENOENT;

	/* An check that it supports classes */
	cops = q->ops->cl_ops;
	if (cops == NULL)
		return -EINVAL;

	/* Now try to get class */
	if (clid == 0) {
		if (pid == TC_H_ROOT)
			clid = qid;
	} else
		clid = TC_H_MAKE(qid, clid);

	if (clid)
		cl = cops->get(q, clid);

	if (cl == 0) {
		err = -ENOENT;
		if (n->nlmsg_type != RTM_NEWTCLASS || !(n->nlmsg_flags&NLM_F_CREATE))
			goto out;
	} else {
		switch (n->nlmsg_type) {
		case RTM_NEWTCLASS:
			err = -EEXIST;
			if (n->nlmsg_flags&NLM_F_EXCL)
				goto out;
			break;
		case RTM_DELTCLASS:
			err = cops->delete(q, cl);
			if (err == 0)
				tclass_notify(skb, n, q, cl, RTM_DELTCLASS);
			goto out;
		case RTM_GETTCLASS:
			err = tclass_notify(skb, n, q, cl, RTM_NEWTCLASS);
			goto out;
		default:
			err = -EINVAL;
			goto out;
		}
	}

	new_cl = cl;
	err = cops->change(q, clid, pid, tca, &new_cl);
	if (err == 0)
		tclass_notify(skb, n, q, new_cl, RTM_NEWTCLASS);

out:
	if (cl)
		cops->put(q, cl);

	return err;
}


static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q,
			  unsigned long cl,
			  u32 pid, u32 seq, u16 flags, int event)
{
	struct tcmsg *tcm;
	struct nlmsghdr  *nlh;
	unsigned char *b = skb_tail_pointer(skb);
	struct gnet_dump d;
	struct Qdisc_class_ops *cl_ops = q->ops->cl_ops;

	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*tcm), flags);
	tcm = NLMSG_DATA(nlh);
	tcm->tcm_family = AF_UNSPEC;
	tcm->tcm_ifindex = q->dev->ifindex;
	tcm->tcm_parent = q->handle;
	tcm->tcm_handle = q->handle;
	tcm->tcm_info = 0;
	RTA_PUT(skb, TCA_KIND, IFNAMSIZ, q->ops->id);
	if (cl_ops->dump && cl_ops->dump(q, cl, skb, tcm) < 0)
		goto rtattr_failure;

	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
			TCA_XSTATS, q->stats_lock, &d) < 0)
		goto rtattr_failure;

	if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0)
		goto rtattr_failure;

	if (gnet_stats_finish_copy(&d) < 0)
		goto rtattr_failure;

	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
	return skb->len;

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

static int tclass_notify(struct sk_buff *oskb, struct nlmsghdr *n,
			  struct Qdisc *q, unsigned long cl, int event)
{
	struct sk_buff *skb;
	u32 pid = oskb ? NETLINK_CB(oskb).pid : 0;

	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
	if (!skb)
		return -ENOBUFS;

	if (tc_fill_tclass(skb, q, cl, pid, n->nlmsg_seq, 0, event) < 0) {
		kfree_skb(skb);
		return -EINVAL;
	}

	return rtnetlink_send(skb, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO);
}

struct qdisc_dump_args
{
	struct qdisc_walker w;
	struct sk_buff *skb;
	struct netlink_callback *cb;
};

static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walker *arg)
{
	struct qdisc_dump_args *a = (struct qdisc_dump_args *)arg;

	return tc_fill_tclass(a->skb, q, cl, NETLINK_CB(a->cb->skb).pid,
			      a->cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWTCLASS);
}

static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
{
	int t;
	int s_t;
	struct net_device *dev;
	struct Qdisc *q;
	struct tcmsg *tcm = (struct tcmsg*)NLMSG_DATA(cb->nlh);
	struct qdisc_dump_args arg;

	if (cb->nlh->nlmsg_len < NLMSG_LENGTH(sizeof(*tcm)))
		return 0;
	if ((dev = dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
		return 0;

	s_t = cb->args[0];
	t = 0;

	list_for_each_entry(q, &dev->qdisc_list, list) {
		if (t < s_t || !q->ops->cl_ops ||
		    (tcm->tcm_parent &&
		     TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
			t++;
			continue;
		}
		if (t > s_t)
			memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0]));
		arg.w.fn = qdisc_class_dump;
		arg.skb = skb;
		arg.cb = cb;
		arg.w.stop  = 0;
		arg.w.skip = cb->args[1];
		arg.w.count = 0;
		q->ops->cl_ops->walk(q, &arg.w);
		cb->args[1] = arg.w.count;
		if (arg.w.stop)
			break;
		t++;
	}

	cb->args[0] = t;

	dev_put(dev);
	return skb->len;
}

/* Main classifier routine: scans classifier chain attached
   to this qdisc, (optionally) tests for protocol and asks
   specific classifiers.
 */
int tc_classify_compat(struct sk_buff *skb, struct tcf_proto *tp,
		       struct tcf_result *res)
{
	__be16 protocol = skb->protocol;
	int err = 0;

	for (; tp; tp = tp->next) {
		if ((tp->protocol == protocol ||
		     tp->protocol == htons(ETH_P_ALL)) &&
		    (err = tp->classify(skb, tp, res)) >= 0) {
#ifdef CONFIG_NET_CLS_ACT
			if (err != TC_ACT_RECLASSIFY && skb->tc_verd)
				skb->tc_verd = SET_TC_VERD(skb->tc_verd, 0);
#endif
			return err;
		}
	}
	return -1;
}
EXPORT_SYMBOL(tc_classify_compat);

int tc_classify(struct sk_buff *skb, struct tcf_proto *tp,
		struct tcf_result *res)
{
	int err = 0;
	__be16 protocol;
#ifdef CONFIG_NET_CLS_ACT
	struct tcf_proto *otp = tp;
reclassify:
#endif
	protocol = skb->protocol;

	err = tc_classify_compat(skb, tp, res);
#ifdef CONFIG_NET_CLS_ACT
	if (err == TC_ACT_RECLASSIFY) {
		u32 verd = G_TC_VERD(skb->tc_verd);
		tp = otp;

		if (verd++ >= MAX_REC_LOOP) {
			printk("rule prio %u protocol %02x reclassify loop, "
			       "packet dropped\n",
			       tp->prio&0xffff, ntohs(tp->protocol));
			return TC_ACT_SHOT;
		}
		skb->tc_verd = SET_TC_VERD(skb->tc_verd, verd);
		goto reclassify;
	}
#endif
	return err;
}
EXPORT_SYMBOL(tc_classify);

void tcf_destroy(struct tcf_proto *tp)
{
	tp->ops->destroy(tp);
	module_put(tp->ops->owner);
	kfree(tp);
}

void tcf_destroy_chain(struct tcf_proto *fl)
{
	struct tcf_proto *tp;

	while ((tp = fl) != NULL) {
		fl = tp->next;
		tcf_destroy(tp);
	}
}
EXPORT_SYMBOL(tcf_destroy_chain);

#ifdef CONFIG_PROC_FS
static int psched_show(struct seq_file *seq, void *v)
{
	struct timespec ts;

	hrtimer_get_res(CLOCK_MONOTONIC, &ts);
	seq_printf(seq, "%08x %08x %08x %08x\n",
		   (u32)NSEC_PER_USEC, (u32)PSCHED_US2NS(1),
		   1000000,
		   (u32)NSEC_PER_SEC/(u32)ktime_to_ns(timespec_to_ktime(ts)));

	return 0;
}

static int psched_open(struct inode *inode, struct file *file)
{
	return single_open(file, psched_show, PDE(inode)->data);
}

static const struct file_operations psched_fops = {
	.owner = THIS_MODULE,
	.open = psched_open,
	.read  = seq_read,
	.llseek = seq_lseek,
	.release = single_release,
};
#endif

static int __init pktsched_init(void)
{
	register_qdisc(&pfifo_qdisc_ops);
	register_qdisc(&bfifo_qdisc_ops);
	proc_net_fops_create(&init_net, "psched", 0, &psched_fops);

	rtnl_register(PF_UNSPEC, RTM_NEWQDISC, tc_modify_qdisc, NULL);
	rtnl_register(PF_UNSPEC, RTM_DELQDISC, tc_get_qdisc, NULL);
	rtnl_register(PF_UNSPEC, RTM_GETQDISC, tc_get_qdisc, tc_dump_qdisc);
	rtnl_register(PF_UNSPEC, RTM_NEWTCLASS, tc_ctl_tclass, NULL);
	rtnl_register(PF_UNSPEC, RTM_DELTCLASS, tc_ctl_tclass, NULL);
	rtnl_register(PF_UNSPEC, RTM_GETTCLASS, tc_ctl_tclass, tc_dump_tclass);

	return 0;
}

subsys_initcall(pktsched_init);

EXPORT_SYMBOL(qdisc_get_rtab);
EXPORT_SYMBOL(qdisc_put_rtab);
EXPORT_SYMBOL(register_qdisc);
EXPORT_SYMBOL(unregister_qdisc);