[mirror_ubuntu-focal-kernel.git] / net / ipv6 / ip6_offload.c

/*
 *	IPV6 GSO/GRO offload support
 *	Linux INET6 implementation
 *
 *	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, or (at your option) any later version.
 */

#include <linux/kernel.h>
#include <linux/socket.h>
#include <linux/netdevice.h>
#include <linux/skbuff.h>
#include <linux/printk.h>

#include <net/protocol.h>
#include <net/ipv6.h>

#include "ip6_offload.h"

static int ipv6_gso_pull_exthdrs(struct sk_buff *skb, int proto)
{
	const struct net_offload *ops = NULL;

	for (;;) {
		struct ipv6_opt_hdr *opth;
		int len;

		if (proto != NEXTHDR_HOP) {
			ops = rcu_dereference(inet6_offloads[proto]);

			if (unlikely(!ops))
				break;

			if (!(ops->flags & INET6_PROTO_GSO_EXTHDR))
				break;
		}

		if (unlikely(!pskb_may_pull(skb, 8)))
			break;

		opth = (void *)skb->data;
		len = ipv6_optlen(opth);

		if (unlikely(!pskb_may_pull(skb, len)))
			break;

		proto = opth->nexthdr;
		__skb_pull(skb, len);
	}

	return proto;
}

static int ipv6_gso_send_check(struct sk_buff *skb)
{
	const struct ipv6hdr *ipv6h;
	const struct net_offload *ops;
	int err = -EINVAL;

	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
		goto out;

	ipv6h = ipv6_hdr(skb);
	__skb_pull(skb, sizeof(*ipv6h));
	err = -EPROTONOSUPPORT;

	ops = rcu_dereference(inet6_offloads[
		ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr)]);

	if (likely(ops && ops->callbacks.gso_send_check)) {
		skb_reset_transport_header(skb);
		err = ops->callbacks.gso_send_check(skb);
	}

out:
	return err;
}

static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb,
	netdev_features_t features)
{
	struct sk_buff *segs = ERR_PTR(-EINVAL);
	struct ipv6hdr *ipv6h;
	const struct net_offload *ops;
	int proto;
	struct frag_hdr *fptr;
	unsigned int unfrag_ip6hlen;
	u8 *prevhdr;
	int offset = 0;
	bool tunnel;
	int nhoff;

	if (unlikely(skb_shinfo(skb)->gso_type &
		     ~(SKB_GSO_UDP |
		       SKB_GSO_DODGY |
		       SKB_GSO_TCP_ECN |
		       SKB_GSO_GRE |
		       SKB_GSO_IPIP |
		       SKB_GSO_SIT |
		       SKB_GSO_UDP_TUNNEL |
		       SKB_GSO_MPLS |
		       SKB_GSO_TCPV6 |
		       0)))
		goto out;

	skb_reset_network_header(skb);
	nhoff = skb_network_header(skb) - skb_mac_header(skb);
	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
		goto out;

	tunnel = SKB_GSO_CB(skb)->encap_level > 0;
	if (tunnel)
		features = skb->dev->hw_enc_features & netif_skb_features(skb);
	SKB_GSO_CB(skb)->encap_level += sizeof(*ipv6h);

	ipv6h = ipv6_hdr(skb);
	__skb_pull(skb, sizeof(*ipv6h));
	segs = ERR_PTR(-EPROTONOSUPPORT);

	proto = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr);

	ops = rcu_dereference(inet6_offloads[proto]);
	if (likely(ops && ops->callbacks.gso_segment)) {
		skb_reset_transport_header(skb);
		segs = ops->callbacks.gso_segment(skb, features);
	}

	if (IS_ERR(segs))
		goto out;

	for (skb = segs; skb; skb = skb->next) {
		ipv6h = (struct ipv6hdr *)(skb_mac_header(skb) + nhoff);
		ipv6h->payload_len = htons(skb->len - nhoff - sizeof(*ipv6h));
		if (tunnel) {
			skb_reset_inner_headers(skb);
			skb->encapsulation = 1;
		}
		skb->network_header = (u8 *)ipv6h - skb->head;

		if (!tunnel && proto == IPPROTO_UDP) {
			unfrag_ip6hlen = ip6_find_1stfragopt(skb, &prevhdr);
			fptr = (struct frag_hdr *)((u8 *)ipv6h + unfrag_ip6hlen);
			fptr->frag_off = htons(offset);
			if (skb->next != NULL)
				fptr->frag_off |= htons(IP6_MF);
			offset += (ntohs(ipv6h->payload_len) -
				   sizeof(struct frag_hdr));
		}
	}

out:
	return segs;
}

static struct sk_buff **ipv6_gro_receive(struct sk_buff **head,
					 struct sk_buff *skb)
{
	const struct net_offload *ops;
	struct sk_buff **pp = NULL;
	struct sk_buff *p;
	struct ipv6hdr *iph;
	unsigned int nlen;
	unsigned int hlen;
	unsigned int off;
	int flush = 1;
	int proto;
	__wsum csum;

	off = skb_gro_offset(skb);
	hlen = off + sizeof(*iph);
	iph = skb_gro_header_fast(skb, off);
	if (skb_gro_header_hard(skb, hlen)) {
		iph = skb_gro_header_slow(skb, hlen, off);
		if (unlikely(!iph))
			goto out;
	}

	skb_gro_pull(skb, sizeof(*iph));
	skb_set_transport_header(skb, skb_gro_offset(skb));

	flush += ntohs(iph->payload_len) != skb_gro_len(skb);

	rcu_read_lock();
	proto = iph->nexthdr;
	ops = rcu_dereference(inet6_offloads[proto]);
	if (!ops || !ops->callbacks.gro_receive) {
		__pskb_pull(skb, skb_gro_offset(skb));
		proto = ipv6_gso_pull_exthdrs(skb, proto);
		skb_gro_pull(skb, -skb_transport_offset(skb));
		skb_reset_transport_header(skb);
		__skb_push(skb, skb_gro_offset(skb));

		ops = rcu_dereference(inet6_offloads[proto]);
		if (!ops || !ops->callbacks.gro_receive)
			goto out_unlock;

		iph = ipv6_hdr(skb);
	}

	NAPI_GRO_CB(skb)->proto = proto;

	flush--;
	nlen = skb_network_header_len(skb);

	for (p = *head; p; p = p->next) {
		const struct ipv6hdr *iph2;
		__be32 first_word; /* <Version:4><Traffic_Class:8><Flow_Label:20> */

		if (!NAPI_GRO_CB(p)->same_flow)
			continue;

		iph2 = ipv6_hdr(p);
		first_word = *(__be32 *)iph ^ *(__be32 *)iph2 ;

		/* All fields must match except length and Traffic Class. */
		if (nlen != skb_network_header_len(p) ||
		    (first_word & htonl(0xF00FFFFF)) ||
		    memcmp(&iph->nexthdr, &iph2->nexthdr,
			   nlen - offsetof(struct ipv6hdr, nexthdr))) {
			NAPI_GRO_CB(p)->same_flow = 0;
			continue;
		}
		/* flush if Traffic Class fields are different */
		NAPI_GRO_CB(p)->flush |= !!(first_word & htonl(0x0FF00000));
		NAPI_GRO_CB(p)->flush |= flush;
	}

	NAPI_GRO_CB(skb)->flush |= flush;

	csum = skb->csum;
	skb_postpull_rcsum(skb, iph, skb_network_header_len(skb));

	pp = ops->callbacks.gro_receive(head, skb);

	skb->csum = csum;

out_unlock:
	rcu_read_unlock();

out:
	NAPI_GRO_CB(skb)->flush |= flush;

	return pp;
}

static int ipv6_gro_complete(struct sk_buff *skb)
{
	const struct net_offload *ops;
	struct ipv6hdr *iph = ipv6_hdr(skb);
	int err = -ENOSYS;

	iph->payload_len = htons(skb->len - skb_network_offset(skb) -
				 sizeof(*iph));

	rcu_read_lock();
	ops = rcu_dereference(inet6_offloads[NAPI_GRO_CB(skb)->proto]);
	if (WARN_ON(!ops || !ops->callbacks.gro_complete))
		goto out_unlock;

	err = ops->callbacks.gro_complete(skb);

out_unlock:
	rcu_read_unlock();

	return err;
}

static struct packet_offload ipv6_packet_offload __read_mostly = {
	.type = cpu_to_be16(ETH_P_IPV6),
	.callbacks = {
		.gso_send_check = ipv6_gso_send_check,
		.gso_segment = ipv6_gso_segment,
		.gro_receive = ipv6_gro_receive,
		.gro_complete = ipv6_gro_complete,
	},
};

static const struct net_offload sit_offload = {
	.callbacks = {
		.gso_send_check = ipv6_gso_send_check,
		.gso_segment	= ipv6_gso_segment,
	},
};

static int __init ipv6_offload_init(void)
{

	if (tcpv6_offload_init() < 0)
		pr_crit("%s: Cannot add TCP protocol offload\n", __func__);
	if (udp_offload_init() < 0)
		pr_crit("%s: Cannot add UDP protocol offload\n", __func__);
	if (ipv6_exthdrs_offload_init() < 0)
		pr_crit("%s: Cannot add EXTHDRS protocol offload\n", __func__);

	dev_add_offload(&ipv6_packet_offload);

	inet_add_offload(&sit_offload, IPPROTO_IPV6);

	return 0;
}

fs_initcall(ipv6_offload_init);
Commit	Line	Data
d1da932e VY	1	/*
	2	* IPV6 GSO/GRO offload support
	3	* Linux INET6 implementation
	4	*
	5	* This program is free software; you can redistribute it and/or
	6	* modify it under the terms of the GNU General Public License
	7	* as published by the Free Software Foundation; either version
	8	* 2 of the License, or (at your option) any later version.
	9	*/
	10
	11	#include <linux/kernel.h>
	12	#include <linux/socket.h>
	13	#include <linux/netdevice.h>
	14	#include <linux/skbuff.h>
c6b641a4	15	#include <linux/printk.h>
d1da932e VY	16
	17	#include <net/protocol.h>
	18	#include <net/ipv6.h>
	19
	20	#include "ip6_offload.h"
	21
	22	static int ipv6_gso_pull_exthdrs(struct sk_buff *skb, int proto)
	23	{
	24	const struct net_offload *ops = NULL;
	25
	26	for (;;) {
	27	struct ipv6_opt_hdr *opth;
	28	int len;
	29
	30	if (proto != NEXTHDR_HOP) {
	31	ops = rcu_dereference(inet6_offloads[proto]);
	32
	33	if (unlikely(!ops))
	34	break;
	35
	36	if (!(ops->flags & INET6_PROTO_GSO_EXTHDR))
	37	break;
	38	}
	39
	40	if (unlikely(!pskb_may_pull(skb, 8)))
	41	break;
	42
	43	opth = (void *)skb->data;
	44	len = ipv6_optlen(opth);
	45
	46	if (unlikely(!pskb_may_pull(skb, len)))
	47	break;
	48
	49	proto = opth->nexthdr;
	50	__skb_pull(skb, len);
	51	}
	52
	53	return proto;
	54	}
	55
	56	static int ipv6_gso_send_check(struct sk_buff *skb)
	57	{
	58	const struct ipv6hdr *ipv6h;
	59	const struct net_offload *ops;
	60	int err = -EINVAL;
	61
	62	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
	63	goto out;
	64
	65	ipv6h = ipv6_hdr(skb);
	66	__skb_pull(skb, sizeof(*ipv6h));
	67	err = -EPROTONOSUPPORT;
	68
d1da932e VY	69	ops = rcu_dereference(inet6_offloads[
	70	ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr)]);
	71
f191a1d1	72	if (likely(ops && ops->callbacks.gso_send_check)) {
d1da932e	73	skb_reset_transport_header(skb);
f191a1d1	74	err = ops->callbacks.gso_send_check(skb);
d1da932e	75	}
d1da932e VY	76
	77	out:
	78	return err;
	79	}
	80
	81	static struct sk_buff ipv6_gso_segment(struct sk_buff skb,
	82	netdev_features_t features)
	83	{
	84	struct sk_buff *segs = ERR_PTR(-EINVAL);
	85	struct ipv6hdr *ipv6h;
	86	const struct net_offload *ops;
	87	int proto;
	88	struct frag_hdr *fptr;
	89	unsigned int unfrag_ip6hlen;
	90	u8 *prevhdr;
	91	int offset = 0;
d949d826	92	bool tunnel;
d3e5e006	93	int nhoff;
d1da932e	94
d1da932e VY	95	if (unlikely(skb_shinfo(skb)->gso_type &
	96	~(SKB_GSO_UDP \|
	97	SKB_GSO_DODGY \|
	98	SKB_GSO_TCP_ECN \|
68c33163	99	SKB_GSO_GRE \|
cb32f511	100	SKB_GSO_IPIP \|
61c1db7f	101	SKB_GSO_SIT \|
73136267	102	SKB_GSO_UDP_TUNNEL \|
0d89d203	103	SKB_GSO_MPLS \|
d1da932e VY	104	SKB_GSO_TCPV6 \|
	105	0)))
	106	goto out;
	107
d3e5e006 ED	108	skb_reset_network_header(skb);
d3e5e006 ED	109	nhoff = skb_network_header(skb) - skb_mac_header(skb);
d1da932e VY	110	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
	111	goto out;
	112
d3e5e006 ED	113	tunnel = SKB_GSO_CB(skb)->encap_level > 0;
	114	if (tunnel)
	115	features = skb->dev->hw_enc_features & netif_skb_features(skb);
	116	SKB_GSO_CB(skb)->encap_level += sizeof(*ipv6h);
	117
d1da932e VY	118	ipv6h = ipv6_hdr(skb);
	119	__skb_pull(skb, sizeof(*ipv6h));
	120	segs = ERR_PTR(-EPROTONOSUPPORT);
	121
	122	proto = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr);
b917eb15	123
d1da932e	124	ops = rcu_dereference(inet6_offloads[proto]);
f191a1d1	125	if (likely(ops && ops->callbacks.gso_segment)) {
d1da932e	126	skb_reset_transport_header(skb);
f191a1d1	127	segs = ops->callbacks.gso_segment(skb, features);
d1da932e	128	}
d1da932e VY	129
	130	if (IS_ERR(segs))
	131	goto out;
	132
	133	for (skb = segs; skb; skb = skb->next) {
d3e5e006 ED	134	ipv6h = (struct ipv6hdr *)(skb_mac_header(skb) + nhoff);
	135	ipv6h->payload_len = htons(skb->len - nhoff - sizeof(*ipv6h));
	136	if (tunnel) {
	137	skb_reset_inner_headers(skb);
	138	skb->encapsulation = 1;
	139	}
	140	skb->network_header = (u8 *)ipv6h - skb->head;
	141
d949d826	142	if (!tunnel && proto == IPPROTO_UDP) {
d1da932e	143	unfrag_ip6hlen = ip6_find_1stfragopt(skb, &prevhdr);
d3e5e006	144	fptr = (struct frag_hdr )((u8 )ipv6h + unfrag_ip6hlen);
d1da932e VY	145	fptr->frag_off = htons(offset);
	146	if (skb->next != NULL)
	147	fptr->frag_off \|= htons(IP6_MF);
	148	offset += (ntohs(ipv6h->payload_len) -
	149	sizeof(struct frag_hdr));
	150	}
	151	}
	152
	153	out:
	154	return segs;
	155	}
	156
	157	static struct sk_buff ipv6_gro_receive(struct sk_buff head,
	158	struct sk_buff *skb)
	159	{
	160	const struct net_offload *ops;
	161	struct sk_buff **pp = NULL;
	162	struct sk_buff *p;
	163	struct ipv6hdr *iph;
	164	unsigned int nlen;
	165	unsigned int hlen;
	166	unsigned int off;
	167	int flush = 1;
	168	int proto;
	169	__wsum csum;
	170
	171	off = skb_gro_offset(skb);
	172	hlen = off + sizeof(*iph);
	173	iph = skb_gro_header_fast(skb, off);
	174	if (skb_gro_header_hard(skb, hlen)) {
	175	iph = skb_gro_header_slow(skb, hlen, off);
	176	if (unlikely(!iph))
	177	goto out;
	178	}
	179
	180	skb_gro_pull(skb, sizeof(*iph));
	181	skb_set_transport_header(skb, skb_gro_offset(skb));
	182
	183	flush += ntohs(iph->payload_len) != skb_gro_len(skb);
	184
	185	rcu_read_lock();
	186	proto = iph->nexthdr;
	187	ops = rcu_dereference(inet6_offloads[proto]);
f191a1d1	188	if (!ops \|\| !ops->callbacks.gro_receive) {
d1da932e VY	189	__pskb_pull(skb, skb_gro_offset(skb));
	190	proto = ipv6_gso_pull_exthdrs(skb, proto);
	191	skb_gro_pull(skb, -skb_transport_offset(skb));
	192	skb_reset_transport_header(skb);
	193	__skb_push(skb, skb_gro_offset(skb));
	194
	195	ops = rcu_dereference(inet6_offloads[proto]);
f191a1d1	196	if (!ops \|\| !ops->callbacks.gro_receive)
d1da932e VY	197	goto out_unlock;
	198
	199	iph = ipv6_hdr(skb);
	200	}
	201
	202	NAPI_GRO_CB(skb)->proto = proto;
	203
	204	flush--;
	205	nlen = skb_network_header_len(skb);
	206
	207	for (p = *head; p; p = p->next) {
	208	const struct ipv6hdr *iph2;
	209	__be32 first_word; /* <Version:4><Traffic_Class:8><Flow_Label:20> */
	210
	211	if (!NAPI_GRO_CB(p)->same_flow)
	212	continue;
	213
	214	iph2 = ipv6_hdr(p);
	215	first_word = (__be32 )iph ^ (__be32 )iph2 ;
	216
	217	/* All fields must match except length and Traffic Class. */
	218	if (nlen != skb_network_header_len(p) \|\|
	219	(first_word & htonl(0xF00FFFFF)) \|\|
	220	memcmp(&iph->nexthdr, &iph2->nexthdr,
	221	nlen - offsetof(struct ipv6hdr, nexthdr))) {
	222	NAPI_GRO_CB(p)->same_flow = 0;
	223	continue;
	224	}
	225	/* flush if Traffic Class fields are different */
	226	NAPI_GRO_CB(p)->flush \|= !!(first_word & htonl(0x0FF00000));
	227	NAPI_GRO_CB(p)->flush \|= flush;
	228	}
	229
	230	NAPI_GRO_CB(skb)->flush \|= flush;
	231
	232	csum = skb->csum;
	233	skb_postpull_rcsum(skb, iph, skb_network_header_len(skb));
	234
f191a1d1	235	pp = ops->callbacks.gro_receive(head, skb);
d1da932e VY	236
	237	skb->csum = csum;
	238
	239	out_unlock:
	240	rcu_read_unlock();
	241
	242	out:
	243	NAPI_GRO_CB(skb)->flush \|= flush;
	244
	245	return pp;
	246	}
	247
	248	static int ipv6_gro_complete(struct sk_buff *skb)
	249	{
	250	const struct net_offload *ops;
	251	struct ipv6hdr *iph = ipv6_hdr(skb);
	252	int err = -ENOSYS;
	253
	254	iph->payload_len = htons(skb->len - skb_network_offset(skb) -
	255	sizeof(*iph));
	256
	257	rcu_read_lock();
	258	ops = rcu_dereference(inet6_offloads[NAPI_GRO_CB(skb)->proto]);
f191a1d1	259	if (WARN_ON(!ops \|\| !ops->callbacks.gro_complete))
d1da932e VY	260	goto out_unlock;
d1da932e VY	261
f191a1d1	262	err = ops->callbacks.gro_complete(skb);
d1da932e VY	263
	264	out_unlock:
	265	rcu_read_unlock();
	266
	267	return err;
	268	}
	269
	270	static struct packet_offload ipv6_packet_offload __read_mostly = {
	271	.type = cpu_to_be16(ETH_P_IPV6),
f191a1d1 VY	272	.callbacks = {
	273	.gso_send_check = ipv6_gso_send_check,
	274	.gso_segment = ipv6_gso_segment,
	275	.gro_receive = ipv6_gro_receive,
	276	.gro_complete = ipv6_gro_complete,
	277	},
d1da932e VY	278	};
d1da932e VY	279
61c1db7f ED	280	static const struct net_offload sit_offload = {
	281	.callbacks = {
	282	.gso_send_check = ipv6_gso_send_check,
	283	.gso_segment = ipv6_gso_segment,
	284	},
	285	};
	286
c6b641a4	287	static int __init ipv6_offload_init(void)
d1da932e	288	{
c6b641a4 VY	289
	290	if (tcpv6_offload_init() < 0)
	291	pr_crit("%s: Cannot add TCP protocol offload\n", __func__);
	292	if (udp_offload_init() < 0)
	293	pr_crit("%s: Cannot add UDP protocol offload\n", __func__);
	294	if (ipv6_exthdrs_offload_init() < 0)
	295	pr_crit("%s: Cannot add EXTHDRS protocol offload\n", __func__);
	296
d1da932e	297	dev_add_offload(&ipv6_packet_offload);
61c1db7f ED	298
	299	inet_add_offload(&sit_offload, IPPROTO_IPV6);
	300
c6b641a4	301	return 0;
d1da932e VY	302	}
d1da932e VY	303
c6b641a4	304	fs_initcall(ipv6_offload_init);