tcp_offload.c 8.34 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12
/*
 *	IPV4 GSO/GRO offload support
 *	Linux INET 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.
 *
 *	TCPv4 GSO/GRO support
 */

13
#include <linux/indirect_call_wrapper.h>
14 15 16 17
#include <linux/skbuff.h>
#include <net/tcp.h>
#include <net/protocol.h>

18 19
static void tcp_gso_tstamp(struct sk_buff *skb, unsigned int ts_seq,
			   unsigned int seq, unsigned int mss)
20 21
{
	while (skb) {
22 23
		if (before(ts_seq, seq + mss)) {
			skb_shinfo(skb)->tx_flags |= SKBTX_SW_TSTAMP;
24 25 26 27 28 29 30 31 32
			skb_shinfo(skb)->tskey = ts_seq;
			return;
		}

		skb = skb->next;
		seq += mss;
	}
}

Eric Dumazet's avatar
Eric Dumazet committed
33 34
static struct sk_buff *tcp4_gso_segment(struct sk_buff *skb,
					netdev_features_t features)
35
{
36 37 38
	if (!(skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4))
		return ERR_PTR(-EINVAL);

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
	if (!pskb_may_pull(skb, sizeof(struct tcphdr)))
		return ERR_PTR(-EINVAL);

	if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL)) {
		const struct iphdr *iph = ip_hdr(skb);
		struct tcphdr *th = tcp_hdr(skb);

		/* Set up checksum pseudo header, usually expect stack to
		 * have done this already.
		 */

		th->check = 0;
		skb->ip_summed = CHECKSUM_PARTIAL;
		__tcp_v4_send_check(skb, iph->saddr, iph->daddr);
	}

	return tcp_gso_segment(skb, features);
}

Eric Dumazet's avatar
Eric Dumazet committed
58
struct sk_buff *tcp_gso_segment(struct sk_buff *skb,
59 60 61
				netdev_features_t features)
{
	struct sk_buff *segs = ERR_PTR(-EINVAL);
62
	unsigned int sum_truesize = 0;
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
	struct tcphdr *th;
	unsigned int thlen;
	unsigned int seq;
	__be32 delta;
	unsigned int oldlen;
	unsigned int mss;
	struct sk_buff *gso_skb = skb;
	__sum16 newcheck;
	bool ooo_okay, copy_destructor;

	th = tcp_hdr(skb);
	thlen = th->doff * 4;
	if (thlen < sizeof(*th))
		goto out;

	if (!pskb_may_pull(skb, thlen))
		goto out;

	oldlen = (u16)~skb->len;
	__skb_pull(skb, thlen);

84
	mss = skb_shinfo(skb)->gso_size;
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
	if (unlikely(skb->len <= mss))
		goto out;

	if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) {
		/* Packet is from an untrusted source, reset gso_segs. */

		skb_shinfo(skb)->gso_segs = DIV_ROUND_UP(skb->len, mss);

		segs = NULL;
		goto out;
	}

	copy_destructor = gso_skb->destructor == tcp_wfree;
	ooo_okay = gso_skb->ooo_okay;
	/* All segments but the first should have ooo_okay cleared */
	skb->ooo_okay = 0;

	segs = skb_segment(skb, features);
	if (IS_ERR(segs))
		goto out;

	/* Only first segment might have ooo_okay set */
	segs->ooo_okay = ooo_okay;

109 110 111 112 113 114 115
	/* GSO partial and frag_list segmentation only requires splitting
	 * the frame into an MSS multiple and possibly a remainder, both
	 * cases return a GSO skb. So update the mss now.
	 */
	if (skb_is_gso(segs))
		mss *= skb_shinfo(segs)->gso_segs;

116 117 118 119 120 121
	delta = htonl(oldlen + (thlen + mss));

	skb = segs;
	th = tcp_hdr(skb);
	seq = ntohl(th->seq);

122 123 124
	if (unlikely(skb_shinfo(gso_skb)->tx_flags & SKBTX_SW_TSTAMP))
		tcp_gso_tstamp(segs, skb_shinfo(gso_skb)->tskey, seq, mss);

125 126 127
	newcheck = ~csum_fold((__force __wsum)((__force u32)th->check +
					       (__force u32)delta));

128
	while (skb->next) {
129 130 131
		th->fin = th->psh = 0;
		th->check = newcheck;

132 133 134
		if (skb->ip_summed == CHECKSUM_PARTIAL)
			gso_reset_checksum(skb, ~th->check);
		else
Tom Herbert's avatar
Tom Herbert committed
135
			th->check = gso_make_checksum(skb, ~th->check);
136 137 138 139 140

		seq += mss;
		if (copy_destructor) {
			skb->destructor = gso_skb->destructor;
			skb->sk = gso_skb->sk;
141
			sum_truesize += skb->truesize;
142 143 144 145 146 147
		}
		skb = skb->next;
		th = tcp_hdr(skb);

		th->seq = htonl(seq);
		th->cwr = 0;
148
	}
149 150 151 152 153 154 155

	/* Following permits TCP Small Queues to work well with GSO :
	 * The callback to TCP stack will be called at the time last frag
	 * is freed at TX completion, and not right now when gso_skb
	 * is freed by GSO engine
	 */
	if (copy_destructor) {
156 157
		int delta;

158 159
		swap(gso_skb->sk, skb->sk);
		swap(gso_skb->destructor, skb->destructor);
160
		sum_truesize += skb->truesize;
161 162 163 164 165 166 167 168
		delta = sum_truesize - gso_skb->truesize;
		/* In some pathological cases, delta can be negative.
		 * We need to either use refcount_add() or refcount_sub_and_test()
		 */
		if (likely(delta >= 0))
			refcount_add(delta, &skb->sk->sk_wmem_alloc);
		else
			WARN_ON_ONCE(refcount_sub_and_test(-delta, &skb->sk->sk_wmem_alloc));
169 170 171 172 173 174 175
	}

	delta = htonl(oldlen + (skb_tail_pointer(skb) -
				skb_transport_header(skb)) +
		      skb->data_len);
	th->check = ~csum_fold((__force __wsum)((__force u32)th->check +
				(__force u32)delta));
176 177 178
	if (skb->ip_summed == CHECKSUM_PARTIAL)
		gso_reset_checksum(skb, ~th->check);
	else
Tom Herbert's avatar
Tom Herbert committed
179
		th->check = gso_make_checksum(skb, ~th->check);
180 181 182 183
out:
	return segs;
}

184
struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb)
185
{
186
	struct sk_buff *pp = NULL;
187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223
	struct sk_buff *p;
	struct tcphdr *th;
	struct tcphdr *th2;
	unsigned int len;
	unsigned int thlen;
	__be32 flags;
	unsigned int mss = 1;
	unsigned int hlen;
	unsigned int off;
	int flush = 1;
	int i;

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

	thlen = th->doff * 4;
	if (thlen < sizeof(*th))
		goto out;

	hlen = off + thlen;
	if (skb_gro_header_hard(skb, hlen)) {
		th = skb_gro_header_slow(skb, hlen, off);
		if (unlikely(!th))
			goto out;
	}

	skb_gro_pull(skb, thlen);

	len = skb_gro_len(skb);
	flags = tcp_flag_word(th);

224
	list_for_each_entry(p, head, list) {
225 226 227 228 229 230 231 232 233 234 235 236
		if (!NAPI_GRO_CB(p)->same_flow)
			continue;

		th2 = tcp_hdr(p);

		if (*(u32 *)&th->source ^ *(u32 *)&th2->source) {
			NAPI_GRO_CB(p)->same_flow = 0;
			continue;
		}

		goto found;
	}
237
	p = NULL;
238 239 240
	goto out_check_final;

found:
241
	/* Include the IP ID check below from the inner most IP hdr */
242
	flush = NAPI_GRO_CB(p)->flush;
243 244 245 246 247 248 249 250
	flush |= (__force int)(flags & TCP_FLAG_CWR);
	flush |= (__force int)((flags ^ tcp_flag_word(th2)) &
		  ~(TCP_FLAG_CWR | TCP_FLAG_FIN | TCP_FLAG_PSH));
	flush |= (__force int)(th->ack_seq ^ th2->ack_seq);
	for (i = sizeof(*th); i < thlen; i += 4)
		flush |= *(u32 *)((u8 *)th + i) ^
			 *(u32 *)((u8 *)th2 + i);

251 252 253 254 255 256 257 258 259 260 261
	/* When we receive our second frame we can made a decision on if we
	 * continue this flow as an atomic flow with a fixed ID or if we use
	 * an incrementing ID.
	 */
	if (NAPI_GRO_CB(p)->flush_id != 1 ||
	    NAPI_GRO_CB(p)->count != 1 ||
	    !NAPI_GRO_CB(p)->is_atomic)
		flush |= NAPI_GRO_CB(p)->flush_id;
	else
		NAPI_GRO_CB(p)->is_atomic = false;

262
	mss = skb_shinfo(p)->gso_size;
263 264 265

	flush |= (len - 1) >= mss;
	flush |= (ntohl(th2->seq) + skb_gro_len(p)) ^ ntohl(th->seq);
266 267 268
#ifdef CONFIG_TLS_DEVICE
	flush |= p->decrypted ^ skb->decrypted;
#endif
269

270
	if (flush || skb_gro_receive(p, skb)) {
271 272 273 274 275 276 277 278 279 280 281 282 283
		mss = 1;
		goto out_check_final;
	}

	tcp_flag_word(th2) |= flags & (TCP_FLAG_FIN | TCP_FLAG_PSH);

out_check_final:
	flush = len < mss;
	flush |= (__force int)(flags & (TCP_FLAG_URG | TCP_FLAG_PSH |
					TCP_FLAG_RST | TCP_FLAG_SYN |
					TCP_FLAG_FIN));

	if (p && (!NAPI_GRO_CB(skb)->same_flow || flush))
284
		pp = p;
285 286

out:
287
	NAPI_GRO_CB(skb)->flush |= (flush != 0);
288 289 290 291 292 293 294 295

	return pp;
}

int tcp_gro_complete(struct sk_buff *skb)
{
	struct tcphdr *th = tcp_hdr(skb);

296
	skb->csum_start = (unsigned char *)th - skb->head;
297 298 299 300 301 302 303 304 305 306 307 308
	skb->csum_offset = offsetof(struct tcphdr, check);
	skb->ip_summed = CHECKSUM_PARTIAL;

	skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;

	if (th->cwr)
		skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN;

	return 0;
}
EXPORT_SYMBOL(tcp_gro_complete);

309 310
INDIRECT_CALLABLE_SCOPE
struct sk_buff *tcp4_gro_receive(struct list_head *head, struct sk_buff *skb)
311
{
312
	/* Don't bother verifying checksum if we're going to flush anyway. */
313 314 315
	if (!NAPI_GRO_CB(skb)->flush &&
	    skb_gro_checksum_validate(skb, IPPROTO_TCP,
				      inet_gro_compute_pseudo)) {
316 317 318 319 320 321 322
		NAPI_GRO_CB(skb)->flush = 1;
		return NULL;
	}

	return tcp_gro_receive(head, skb);
}

323
INDIRECT_CALLABLE_SCOPE int tcp4_gro_complete(struct sk_buff *skb, int thoff)
324 325 326 327
{
	const struct iphdr *iph = ip_hdr(skb);
	struct tcphdr *th = tcp_hdr(skb);

328 329
	th->check = ~tcp_v4_check(skb->len - thoff, iph->saddr,
				  iph->daddr, 0);
330
	skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV4;
331

332 333 334
	if (NAPI_GRO_CB(skb)->is_atomic)
		skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_FIXEDID;

335 336 337 338 339
	return tcp_gro_complete(skb);
}

static const struct net_offload tcpv4_offload = {
	.callbacks = {
340
		.gso_segment	=	tcp4_gso_segment,
341 342 343 344 345 346 347 348 349
		.gro_receive	=	tcp4_gro_receive,
		.gro_complete	=	tcp4_gro_complete,
	},
};

int __init tcpv4_offload_init(void)
{
	return inet_add_offload(&tcpv4_offload, IPPROTO_TCP);
}