/* called from BPF program, therefore rcu_read_lock is held
 * bpf_check() verified that 'buf' pointer to BPF's stack
 * and it has 'len' bytes for us to read
 */
void bpf_channel_push_struct(struct bpf_context *pctx, u32 struct_id,
			     const void *buf, u32 len)
{
	struct bpf_dp_context *ctx = container_of(pctx, struct bpf_dp_context,
						  context);
	struct dp_upcall_info upcall;
	struct plum *plum;
	struct nlattr *nla;

	if (unlikely(!ctx->skb))
		return;

	plum = rcu_dereference(ctx->dp->plums[pctx->plum_id]);
	if (unlikely(!plum))
		return;

	/* allocate temp nlattr to pass it into ovs_dp_upcall */
	nla = kzalloc(nla_total_size(4 + len), GFP_ATOMIC);
	if (unlikely(!nla))
		return;

	nla->nla_type = OVS_PACKET_ATTR_USERDATA;
	nla->nla_len = nla_attr_size(4 + len);
	memcpy(nla_data(nla), &struct_id, 4);
	memcpy(nla_data(nla) + 4, buf, len);

	upcall.cmd = OVS_PACKET_CMD_ACTION;
	upcall.key = NULL;
	upcall.userdata = nla;
	upcall.portid = plum->upcall_pid;
	ovs_dp_upcall(ctx->dp, NULL, &upcall);
	kfree(nla);
}
Example #2
0
static int sfq_change(struct Qdisc *sch, struct nlattr *opt)
{
	struct sfq_sched_data *q = qdisc_priv(sch);
	struct tc_sfq_qopt *ctl = nla_data(opt);
	unsigned int qlen;

	if (opt->nla_len < nla_attr_size(sizeof(*ctl)))
		return -EINVAL;

	sch_tree_lock(sch);
	q->quantum = ctl->quantum ? : psched_mtu(sch->dev);
	q->perturb_period = ctl->perturb_period * HZ;
	if (ctl->limit)
		q->limit = min_t(u32, ctl->limit, SFQ_DEPTH - 1);

	qlen = sch->q.qlen;
	while (sch->q.qlen > q->limit)
		sfq_drop(sch);
	qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);

	del_timer(&q->perturb_timer);
	if (q->perturb_period) {
		mod_timer(&q->perturb_timer, jiffies + q->perturb_period);
		q->perturbation = net_random();
	}
	sch_tree_unlock(sch);
	return 0;
}
Example #3
0
static int parse_attr(struct nlattr *tb[], int maxtype, struct nlattr *nla,
		      const struct nla_policy *policy, int len)
{
	int nested_len = nla_len(nla) - NLA_ALIGN(len);

	if (nested_len < 0)
		return -EINVAL;
	if (nested_len >= nla_attr_size(0))
		return nla_parse(tb, maxtype, nla_data(nla) + NLA_ALIGN(len),
				 nested_len, policy);
	memset(tb, 0, sizeof(struct nlattr *) * (maxtype + 1));
	return 0;
}
Example #4
0
static int sfq_change(struct Qdisc *sch, struct nlattr *opt)
{
	struct sfq_sched_data *q = qdisc_priv(sch);
	struct tc_sfq_qopt *ctl = nla_data(opt);
	struct tc_sfq_qopt_v1 *ctl_v1 = NULL;
	unsigned int qlen;
	struct red_parms *p = NULL;

	if (opt->nla_len < nla_attr_size(sizeof(*ctl)))
		return -EINVAL;
	if (opt->nla_len >= nla_attr_size(sizeof(*ctl_v1)))
		ctl_v1 = nla_data(opt);
	if (ctl->divisor &&
	    (!is_power_of_2(ctl->divisor) || ctl->divisor > 65536))
		return -EINVAL;
	if (ctl_v1 && ctl_v1->qth_min) {
		p = kmalloc(sizeof(*p), GFP_KERNEL);
		if (!p)
			return -ENOMEM;
	}
	sch_tree_lock(sch);
	if (ctl->quantum) {
		q->quantum = ctl->quantum;
		q->scaled_quantum = SFQ_ALLOT_SIZE(q->quantum);
	}
	q->perturb_period = ctl->perturb_period * HZ;
	if (ctl->flows)
		q->maxflows = min_t(u32, ctl->flows, SFQ_MAX_FLOWS);
	if (ctl->divisor) {
		q->divisor = ctl->divisor;
		q->maxflows = min_t(u32, q->maxflows, q->divisor);
	}
	if (ctl_v1) {
		if (ctl_v1->depth)
			q->maxdepth = min_t(u32, ctl_v1->depth, SFQ_MAX_DEPTH);
		if (p) {
			swap(q->red_parms, p);
			red_set_parms(q->red_parms,
				      ctl_v1->qth_min, ctl_v1->qth_max,
				      ctl_v1->Wlog,
				      ctl_v1->Plog, ctl_v1->Scell_log,
				      NULL,
				      ctl_v1->max_P);
		}
		q->flags = ctl_v1->flags;
		q->headdrop = ctl_v1->headdrop;
	}
	if (ctl->limit) {
		q->limit = min_t(u32, ctl->limit, q->maxdepth * q->maxflows);
		q->maxflows = min_t(u32, q->maxflows, q->limit);
	}

	qlen = sch->q.qlen;
	while (sch->q.qlen > q->limit)
		sfq_drop(sch);
	qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);

	del_timer(&q->perturb_timer);
	if (q->perturb_period) {
		mod_timer(&q->perturb_timer, jiffies + q->perturb_period);
		q->perturbation = prandom_u32();
	}
	sch_tree_unlock(sch);
	kfree(p);
	return 0;
}
Example #5
0
/* This is an inline function, we don't really care about a long
 * list of arguments */
static inline int
__build_packet_message(struct nfnl_log_net *log,
			struct nfulnl_instance *inst,
			const struct sk_buff *skb,
			unsigned int data_len,
			u_int8_t pf,
			unsigned int hooknum,
			const struct net_device *indev,
			const struct net_device *outdev,
			const char *prefix, unsigned int plen,
			const struct nfnl_ct_hook *nfnl_ct,
			struct nf_conn *ct, enum ip_conntrack_info ctinfo)
{
	struct nfulnl_msg_packet_hdr pmsg;
	struct nlmsghdr *nlh;
	struct nfgenmsg *nfmsg;
	sk_buff_data_t old_tail = inst->skb->tail;
	struct sock *sk;
	const unsigned char *hwhdrp;

	nlh = nlmsg_put(inst->skb, 0, 0,
			nfnl_msg_type(NFNL_SUBSYS_ULOG, NFULNL_MSG_PACKET),
			sizeof(struct nfgenmsg), 0);
	if (!nlh)
		return -1;
	nfmsg = nlmsg_data(nlh);
	nfmsg->nfgen_family = pf;
	nfmsg->version = NFNETLINK_V0;
	nfmsg->res_id = htons(inst->group_num);

	memset(&pmsg, 0, sizeof(pmsg));
	pmsg.hw_protocol	= skb->protocol;
	pmsg.hook		= hooknum;

	if (nla_put(inst->skb, NFULA_PACKET_HDR, sizeof(pmsg), &pmsg))
		goto nla_put_failure;

	if (prefix &&
	    nla_put(inst->skb, NFULA_PREFIX, plen, prefix))
		goto nla_put_failure;

	if (indev) {
#if !IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
		if (nla_put_be32(inst->skb, NFULA_IFINDEX_INDEV,
				 htonl(indev->ifindex)))
			goto nla_put_failure;
#else
		if (pf == PF_BRIDGE) {
			/* Case 1: outdev is physical input device, we need to
			 * look for bridge group (when called from
			 * netfilter_bridge) */
			if (nla_put_be32(inst->skb, NFULA_IFINDEX_PHYSINDEV,
					 htonl(indev->ifindex)) ||
			/* this is the bridge group "brX" */
			/* rcu_read_lock()ed by nf_hook_thresh or
			 * nf_log_packet.
			 */
			    nla_put_be32(inst->skb, NFULA_IFINDEX_INDEV,
					 htonl(br_port_get_rcu(indev)->br->dev->ifindex)))
				goto nla_put_failure;
		} else {
			struct net_device *physindev;

			/* Case 2: indev is bridge group, we need to look for
			 * physical device (when called from ipv4) */
			if (nla_put_be32(inst->skb, NFULA_IFINDEX_INDEV,
					 htonl(indev->ifindex)))
				goto nla_put_failure;

			physindev = nf_bridge_get_physindev(skb);
			if (physindev &&
			    nla_put_be32(inst->skb, NFULA_IFINDEX_PHYSINDEV,
					 htonl(physindev->ifindex)))
				goto nla_put_failure;
		}
#endif
	}

	if (outdev) {
#if !IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
		if (nla_put_be32(inst->skb, NFULA_IFINDEX_OUTDEV,
				 htonl(outdev->ifindex)))
			goto nla_put_failure;
#else
		if (pf == PF_BRIDGE) {
			/* Case 1: outdev is physical output device, we need to
			 * look for bridge group (when called from
			 * netfilter_bridge) */
			if (nla_put_be32(inst->skb, NFULA_IFINDEX_PHYSOUTDEV,
					 htonl(outdev->ifindex)) ||
			/* this is the bridge group "brX" */
			/* rcu_read_lock()ed by nf_hook_thresh or
			 * nf_log_packet.
			 */
			    nla_put_be32(inst->skb, NFULA_IFINDEX_OUTDEV,
					 htonl(br_port_get_rcu(outdev)->br->dev->ifindex)))
				goto nla_put_failure;
		} else {
			struct net_device *physoutdev;

			/* Case 2: indev is a bridge group, we need to look
			 * for physical device (when called from ipv4) */
			if (nla_put_be32(inst->skb, NFULA_IFINDEX_OUTDEV,
					 htonl(outdev->ifindex)))
				goto nla_put_failure;

			physoutdev = nf_bridge_get_physoutdev(skb);
			if (physoutdev &&
			    nla_put_be32(inst->skb, NFULA_IFINDEX_PHYSOUTDEV,
					 htonl(physoutdev->ifindex)))
				goto nla_put_failure;
		}
#endif
	}

	if (skb->mark &&
	    nla_put_be32(inst->skb, NFULA_MARK, htonl(skb->mark)))
		goto nla_put_failure;

	if (indev && skb->dev &&
	    skb->mac_header != skb->network_header) {
		struct nfulnl_msg_packet_hw phw;
		int len;

		memset(&phw, 0, sizeof(phw));
		len = dev_parse_header(skb, phw.hw_addr);
		if (len > 0) {
			phw.hw_addrlen = htons(len);
			if (nla_put(inst->skb, NFULA_HWADDR, sizeof(phw), &phw))
				goto nla_put_failure;
		}
	}

	if (indev && skb_mac_header_was_set(skb)) {
		if (nla_put_be16(inst->skb, NFULA_HWTYPE, htons(skb->dev->type)) ||
		    nla_put_be16(inst->skb, NFULA_HWLEN,
				 htons(skb->dev->hard_header_len)))
			goto nla_put_failure;

		hwhdrp = skb_mac_header(skb);

		if (skb->dev->type == ARPHRD_SIT)
			hwhdrp -= ETH_HLEN;

		if (hwhdrp >= skb->head &&
		    nla_put(inst->skb, NFULA_HWHEADER,
			    skb->dev->hard_header_len, hwhdrp))
			goto nla_put_failure;
	}

	if (skb->tstamp) {
		struct nfulnl_msg_packet_timestamp ts;
		struct timespec64 kts = ktime_to_timespec64(skb->tstamp);
		ts.sec = cpu_to_be64(kts.tv_sec);
		ts.usec = cpu_to_be64(kts.tv_nsec / NSEC_PER_USEC);

		if (nla_put(inst->skb, NFULA_TIMESTAMP, sizeof(ts), &ts))
			goto nla_put_failure;
	}

	/* UID */
	sk = skb->sk;
	if (sk && sk_fullsock(sk)) {
		read_lock_bh(&sk->sk_callback_lock);
		if (sk->sk_socket && sk->sk_socket->file) {
			struct file *file = sk->sk_socket->file;
			const struct cred *cred = file->f_cred;
			struct user_namespace *user_ns = inst->peer_user_ns;
			__be32 uid = htonl(from_kuid_munged(user_ns, cred->fsuid));
			__be32 gid = htonl(from_kgid_munged(user_ns, cred->fsgid));
			read_unlock_bh(&sk->sk_callback_lock);
			if (nla_put_be32(inst->skb, NFULA_UID, uid) ||
			    nla_put_be32(inst->skb, NFULA_GID, gid))
				goto nla_put_failure;
		} else
			read_unlock_bh(&sk->sk_callback_lock);
	}

	/* local sequence number */
	if ((inst->flags & NFULNL_CFG_F_SEQ) &&
	    nla_put_be32(inst->skb, NFULA_SEQ, htonl(inst->seq++)))
		goto nla_put_failure;

	/* global sequence number */
	if ((inst->flags & NFULNL_CFG_F_SEQ_GLOBAL) &&
	    nla_put_be32(inst->skb, NFULA_SEQ_GLOBAL,
			 htonl(atomic_inc_return(&log->global_seq))))
		goto nla_put_failure;

	if (ct && nfnl_ct->build(inst->skb, ct, ctinfo,
				 NFULA_CT, NFULA_CT_INFO) < 0)
		goto nla_put_failure;

	if (data_len) {
		struct nlattr *nla;
		int size = nla_attr_size(data_len);

		if (skb_tailroom(inst->skb) < nla_total_size(data_len))
			goto nla_put_failure;

		nla = skb_put(inst->skb, nla_total_size(data_len));
		nla->nla_type = NFULA_PAYLOAD;
		nla->nla_len = size;

		if (skb_copy_bits(skb, 0, nla_data(nla), data_len))
			BUG();
	}

	nlh->nlmsg_len = inst->skb->tail - old_tail;
	return 0;

nla_put_failure:
	PRINTR(KERN_ERR "nfnetlink_log: error creating log nlmsg\n");
	return -1;
}
Example #6
0
/* This is an inline function, we don't really care about a long
 * list of arguments */
static inline int
__build_packet_message(struct nfulnl_instance *inst,
			const struct sk_buff *skb,
			unsigned int data_len,
			u_int8_t pf,
			unsigned int hooknum,
			const struct net_device *indev,
			const struct net_device *outdev,
			const char *prefix, unsigned int plen)
{
	struct nfulnl_msg_packet_hdr pmsg;
	struct nlmsghdr *nlh;
	struct nfgenmsg *nfmsg;
	sk_buff_data_t old_tail = inst->skb->tail;

	nlh = NLMSG_PUT(inst->skb, 0, 0,
			NFNL_SUBSYS_ULOG << 8 | NFULNL_MSG_PACKET,
			sizeof(struct nfgenmsg));
	nfmsg = NLMSG_DATA(nlh);
	nfmsg->nfgen_family = pf;
	nfmsg->version = NFNETLINK_V0;
	nfmsg->res_id = htons(inst->group_num);

	pmsg.hw_protocol	= skb->protocol;
	pmsg.hook		= hooknum;

	NLA_PUT(inst->skb, NFULA_PACKET_HDR, sizeof(pmsg), &pmsg);

	if (prefix)
		NLA_PUT(inst->skb, NFULA_PREFIX, plen, prefix);

	if (indev) {
#ifndef CONFIG_BRIDGE_NETFILTER
		NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_INDEV,
			     htonl(indev->ifindex));
#else
		if (pf == PF_BRIDGE) {
			/* Case 1: outdev is physical input device, we need to
			 * look for bridge group (when called from
			 * netfilter_bridge) */
			NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_PHYSINDEV,
				     htonl(indev->ifindex));
			/* this is the bridge group "brX" */
			/* rcu_read_lock()ed by nf_hook_slow or nf_log_packet */
			NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_INDEV,
				     htonl(br_port_get_rcu(indev)->br->dev->ifindex));
		} else {
			/* Case 2: indev is bridge group, we need to look for
			 * physical device (when called from ipv4) */
			NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_INDEV,
				     htonl(indev->ifindex));
			if (skb->nf_bridge && skb->nf_bridge->physindev)
				NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_PHYSINDEV,
					     htonl(skb->nf_bridge->physindev->ifindex));
		}
#endif
	}

	if (outdev) {
#ifndef CONFIG_BRIDGE_NETFILTER
		NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_OUTDEV,
			     htonl(outdev->ifindex));
#else
		if (pf == PF_BRIDGE) {
			/* Case 1: outdev is physical output device, we need to
			 * look for bridge group (when called from
			 * netfilter_bridge) */
			NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_PHYSOUTDEV,
				     htonl(outdev->ifindex));
			/* this is the bridge group "brX" */
			/* rcu_read_lock()ed by nf_hook_slow or nf_log_packet */
			NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_OUTDEV,
				     htonl(br_port_get_rcu(outdev)->br->dev->ifindex));
		} else {
			/* Case 2: indev is a bridge group, we need to look
			 * for physical device (when called from ipv4) */
			NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_OUTDEV,
				     htonl(outdev->ifindex));
			if (skb->nf_bridge && skb->nf_bridge->physoutdev)
				NLA_PUT_BE32(inst->skb, NFULA_IFINDEX_PHYSOUTDEV,
					     htonl(skb->nf_bridge->physoutdev->ifindex));
		}
#endif
	}

	if (skb->mark)
		NLA_PUT_BE32(inst->skb, NFULA_MARK, htonl(skb->mark));

	if (indev && skb->dev &&
	    skb->mac_header != skb->network_header) {
		struct nfulnl_msg_packet_hw phw;
		int len = dev_parse_header(skb, phw.hw_addr);
		if (len > 0) {
			phw.hw_addrlen = htons(len);
			NLA_PUT(inst->skb, NFULA_HWADDR, sizeof(phw), &phw);
		}
	}

	if (indev && skb_mac_header_was_set(skb)) {
		NLA_PUT_BE16(inst->skb, NFULA_HWTYPE, htons(skb->dev->type));
		NLA_PUT_BE16(inst->skb, NFULA_HWLEN,
			     htons(skb->dev->hard_header_len));
		NLA_PUT(inst->skb, NFULA_HWHEADER, skb->dev->hard_header_len,
			skb_mac_header(skb));
	}

	if (skb->tstamp.tv64) {
		struct nfulnl_msg_packet_timestamp ts;
		struct timeval tv = ktime_to_timeval(skb->tstamp);
		ts.sec = cpu_to_be64(tv.tv_sec);
		ts.usec = cpu_to_be64(tv.tv_usec);

		NLA_PUT(inst->skb, NFULA_TIMESTAMP, sizeof(ts), &ts);
	}

	/* UID */
	if (skb->sk) {
		read_lock_bh(&skb->sk->sk_callback_lock);
		if (skb->sk->sk_socket && skb->sk->sk_socket->file) {
			struct file *file = skb->sk->sk_socket->file;
			__be32 uid = htonl(file->f_cred->fsuid);
			__be32 gid = htonl(file->f_cred->fsgid);
			/* need to unlock here since NLA_PUT may goto */
			read_unlock_bh(&skb->sk->sk_callback_lock);
			NLA_PUT_BE32(inst->skb, NFULA_UID, uid);
			NLA_PUT_BE32(inst->skb, NFULA_GID, gid);
		} else
			read_unlock_bh(&skb->sk->sk_callback_lock);
	}

	/* local sequence number */
	if (inst->flags & NFULNL_CFG_F_SEQ)
		NLA_PUT_BE32(inst->skb, NFULA_SEQ, htonl(inst->seq++));

	/* global sequence number */
	if (inst->flags & NFULNL_CFG_F_SEQ_GLOBAL)
		NLA_PUT_BE32(inst->skb, NFULA_SEQ_GLOBAL,
			     htonl(atomic_inc_return(&global_seq)));

	if (data_len) {
		struct nlattr *nla;
		int size = nla_attr_size(data_len);

		if (skb_tailroom(inst->skb) < nla_total_size(data_len)) {
			printk(KERN_WARNING "nfnetlink_log: no tailroom!\n");
			goto nlmsg_failure;
		}

		nla = (struct nlattr *)skb_put(inst->skb, nla_total_size(data_len));
		nla->nla_type = NFULA_PAYLOAD;
		nla->nla_len = size;

		if (skb_copy_bits(skb, 0, nla_data(nla), data_len))
			BUG();
	}

	nlh->nlmsg_len = inst->skb->tail - old_tail;
	return 0;

nlmsg_failure:
nla_put_failure:
	PRINTR(KERN_ERR "nfnetlink_log: error creating log nlmsg\n");
	return -1;
}
Example #7
0
static int queue_userspace_packet(int dp_ifindex, struct sk_buff *skb,
				  const struct dp_upcall_info *upcall_info)
{
	struct ovs_header *upcall;
	struct sk_buff *nskb = NULL;
	struct sk_buff *user_skb; /* to be queued to userspace */
	struct nlattr *nla;
	unsigned int len;
	int err;

	if (vlan_tx_tag_present(skb)) {
		nskb = skb_clone(skb, GFP_ATOMIC);
		if (!nskb)
			return -ENOMEM;

		nskb = __vlan_put_tag(nskb, vlan_tx_tag_get(nskb));
		if (!nskb)
			return -ENOMEM;

		nskb->vlan_tci = 0;
		skb = nskb;
	}

	if (nla_attr_size(skb->len) > USHRT_MAX) {
		err = -EFBIG;
		goto out;
	}

	len = sizeof(struct ovs_header);
	len += nla_total_size(skb->len);
	len += nla_total_size(FLOW_BUFSIZE);
	if (upcall_info->cmd == OVS_PACKET_CMD_ACTION)
		len += nla_total_size(8);

	user_skb = genlmsg_new(len, GFP_ATOMIC);
	if (!user_skb) {
		err = -ENOMEM;
		goto out;
	}

	upcall = genlmsg_put(user_skb, 0, 0, &dp_packet_genl_family,
			     0, upcall_info->cmd);
	upcall->dp_ifindex = dp_ifindex;

	nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_KEY);
	ovs_flow_to_nlattrs(upcall_info->key, user_skb);
	nla_nest_end(user_skb, nla);

	if (upcall_info->userdata)
		nla_put_u64(user_skb, OVS_PACKET_ATTR_USERDATA,
			    nla_get_u64(upcall_info->userdata));

	nla = __nla_reserve(user_skb, OVS_PACKET_ATTR_PACKET, skb->len);

	skb_copy_and_csum_dev(skb, nla_data(nla));

	err = genlmsg_unicast(&init_net, user_skb, upcall_info->pid);

out:
	kfree_skb(nskb);
	return err;
}
Example #8
0
/**
 * Return length of padding at the tail of the attribute.
 * @arg payload		Payload length of attribute.
 *
 * @code
 *   +------------------+- - -+- - - - - - - - - +- - -+
 *   | Attribute Header | Pad |     Payload      | Pad |
 *   +------------------+- - -+- - - - - - - - - +- - -+
 *                                                <--->  
 * @endcode
 *
 * @return Length of padding in bytes.
 */
int nla_padlen(int payload)
{
	return nla_total_size(payload) - nla_attr_size(payload);
}
Example #9
0
/**
 * Return size of attribute including padding.
 * @arg payload		Payload length of attribute.
 *
 * @code
 *    <----------- nla_total_size(payload) ----------->
 *   +------------------+- - -+- - - - - - - - - +- - -+
 *   | Attribute Header | Pad |     Payload      | Pad |
 *   +------------------+- - -+- - - - - - - - - +- - -+
 * @endcode
 *
 * @return Size of attribute in bytes.
 */
int nla_total_size(int payload)
{
	return NLA_ALIGN(nla_attr_size(payload));
}
Example #10
0
static int
nm_os_catch_qdisc(struct netmap_generic_adapter *gna, int intercept)
{
	struct netmap_adapter *na = &gna->up.up;
	struct ifnet *ifp = netmap_generic_getifp(gna);
	struct nm_generic_qdisc *qdiscopt = NULL;
	struct Qdisc *fqdisc = NULL;
	struct nlattr *nla = NULL;
	struct netdev_queue *txq;
	unsigned int i;

	if (!gna->txqdisc) {
		return 0;
	}

	if (intercept) {
		nla = kmalloc(nla_attr_size(sizeof(*qdiscopt)),
				GFP_KERNEL);
		if (!nla) {
			D("Failed to allocate netlink attribute");
			return ENOMEM;
		}
		nla->nla_type = RTM_NEWQDISC;
		nla->nla_len = nla_attr_size(sizeof(*qdiscopt));
		qdiscopt = (struct nm_generic_qdisc *)nla_data(nla);
		memset(qdiscopt, 0, sizeof(*qdiscopt));
		qdiscopt->limit = na->num_tx_desc;
	}

	if (ifp->flags & IFF_UP) {
		dev_deactivate(ifp);
	}

	/* Replace the current qdiscs with our own. */
	for (i = 0; i < ifp->real_num_tx_queues; i++) {
		struct Qdisc *nqdisc = NULL;
		struct Qdisc *oqdisc;
		int err;

		txq = netdev_get_tx_queue(ifp, i);

		if (intercept) {
			/* This takes a refcount to netmap module, alloc the
			 * qdisc and calls the init() op with NULL netlink
			 * attribute. */
			nqdisc = qdisc_create_dflt(
#ifndef NETMAP_LINUX_QDISC_CREATE_DFLT_3ARGS
					ifp,
#endif  /* NETMAP_LINUX_QDISC_CREATE_DFLT_3ARGS */
					txq, &generic_qdisc_ops,
					TC_H_UNSPEC);
			if (!nqdisc) {
				D("Failed to create qdisc");
				goto qdisc_create;
			}
			fqdisc = fqdisc ?: nqdisc;

			/* Call the change() op passing a valid netlink
			 * attribute. This is used to set the queue idx. */
			qdiscopt->qidx = i;
			err = nqdisc->ops->change(nqdisc, nla);
			if (err) {
				D("Failed to init qdisc");
				goto qdisc_create;
			}
		}

		oqdisc = dev_graft_qdisc(txq, nqdisc);
		/* We can call this also with
		 * odisc == &noop_qdisc, since the noop
		 * qdisc has the TCQ_F_BUILTIN flag set,
		 * and so qdisc_destroy will skip it. */
		qdisc_destroy(oqdisc);
	}

	kfree(nla);

	if (ifp->qdisc) {
		qdisc_destroy(ifp->qdisc);
	}
	if (intercept) {
		atomic_inc(&fqdisc->refcnt);
		ifp->qdisc = fqdisc;
	} else {
		ifp->qdisc = &noop_qdisc;
	}

	if (ifp->flags & IFF_UP) {
		dev_activate(ifp);
	}

	return 0;

qdisc_create:
	if (nla) {
		kfree(nla);
	}

	nm_os_catch_qdisc(gna, 0);

	return -1;
}
Example #11
0
static int fifo_dump(struct Qdisc *sch, struct sk_buff *skb)
{
	struct tc_fifo_qopt opt = { .limit = sch->limit };

	if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt))
		goto nla_put_failure;
	return skb->len;

nla_put_failure:
	return -1;
}

struct Qdisc_ops pfifo_qdisc_ops __read_mostly = {
	.id		=	"pfifo",
	.priv_size	=	0,
	.enqueue	=	pfifo_enqueue,
	.dequeue	=	qdisc_dequeue_head,
	.peek		=	qdisc_peek_head,
	.init		=	fifo_init,
	.reset		=	qdisc_reset_queue,
	.change		=	fifo_init,
	.dump		=	fifo_dump,
	.owner		=	THIS_MODULE,
};
EXPORT_SYMBOL(pfifo_qdisc_ops);

struct Qdisc_ops bfifo_qdisc_ops __read_mostly = {
	.id		=	"bfifo",
	.priv_size	=	0,
	.enqueue	=	bfifo_enqueue,
	.dequeue	=	qdisc_dequeue_head,
	.peek		=	qdisc_peek_head,
	.init		=	fifo_init,
	.reset		=	qdisc_reset_queue,
	.change		=	fifo_init,
	.dump		=	fifo_dump,
	.owner		=	THIS_MODULE,
};
EXPORT_SYMBOL(bfifo_qdisc_ops);

struct Qdisc_ops pfifo_head_drop_qdisc_ops __read_mostly = {
	.id		=	"pfifo_head_drop",
	.priv_size	=	0,
	.enqueue	=	pfifo_tail_enqueue,
	.dequeue	=	qdisc_dequeue_head,
	.peek		=	qdisc_peek_head,
	.init		=	fifo_init,
	.reset		=	qdisc_reset_queue,
	.change		=	fifo_init,
	.dump		=	fifo_dump,
	.owner		=	THIS_MODULE,
};

/* Pass size change message down to embedded FIFO */
int fifo_set_limit(struct Qdisc *q, unsigned int limit)
{
	struct nlattr *nla;
	int ret = -ENOMEM;

	/* Hack to avoid sending change message to non-FIFO */
	if (strncmp(q->ops->id + 1, "fifo", 4) != 0)
		return 0;

	nla = kmalloc(nla_attr_size(sizeof(struct tc_fifo_qopt)), GFP_KERNEL);
	if (nla) {
		nla->nla_type = RTM_NEWQDISC;
		nla->nla_len = nla_attr_size(sizeof(struct tc_fifo_qopt));
		((struct tc_fifo_qopt *)nla_data(nla))->limit = limit;

		ret = q->ops->change(q, nla);
		kfree(nla);
	}
	return ret;
}
EXPORT_SYMBOL(fifo_set_limit);

struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops,
			       unsigned int limit)
{
	struct Qdisc *q;
	int err = -ENOMEM;

	q = qdisc_create_dflt(sch->dev_queue, ops, TC_H_MAKE(sch->handle, 1));
	if (q) {
		err = fifo_set_limit(q, limit);
		if (err < 0) {
			qdisc_destroy(q);
			q = NULL;
		}
	}

	return q ? : ERR_PTR(err);
}
EXPORT_SYMBOL(fifo_create_dflt);