skmsg.c 28.2 KB
Newer Older
1 2 3 4 5 6 7 8 9
// SPDX-License-Identifier: GPL-2.0
/* Copyright (c) 2017 - 2018 Covalent IO, Inc. http://covalent.io */

#include <linux/skmsg.h>
#include <linux/skbuff.h>
#include <linux/scatterlist.h>

#include <net/sock.h>
#include <net/tcp.h>
10
#include <net/tls.h>
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

static bool sk_msg_try_coalesce_ok(struct sk_msg *msg, int elem_first_coalesce)
{
	if (msg->sg.end > msg->sg.start &&
	    elem_first_coalesce < msg->sg.end)
		return true;

	if (msg->sg.end < msg->sg.start &&
	    (elem_first_coalesce > msg->sg.start ||
	     elem_first_coalesce < msg->sg.end))
		return true;

	return false;
}

int sk_msg_alloc(struct sock *sk, struct sk_msg *msg, int len,
		 int elem_first_coalesce)
{
	struct page_frag *pfrag = sk_page_frag(sk);
30
	u32 osize = msg->sg.size;
31 32 33 34 35 36 37 38
	int ret = 0;

	len -= msg->sg.size;
	while (len > 0) {
		struct scatterlist *sge;
		u32 orig_offset;
		int use, i;

39 40 41 42
		if (!sk_page_frag_refill(sk, pfrag)) {
			ret = -ENOMEM;
			goto msg_trim;
		}
43 44 45

		orig_offset = pfrag->offset;
		use = min_t(int, len, pfrag->size - orig_offset);
46 47 48 49
		if (!sk_wmem_schedule(sk, use)) {
			ret = -ENOMEM;
			goto msg_trim;
		}
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78

		i = msg->sg.end;
		sk_msg_iter_var_prev(i);
		sge = &msg->sg.data[i];

		if (sk_msg_try_coalesce_ok(msg, elem_first_coalesce) &&
		    sg_page(sge) == pfrag->page &&
		    sge->offset + sge->length == orig_offset) {
			sge->length += use;
		} else {
			if (sk_msg_full(msg)) {
				ret = -ENOSPC;
				break;
			}

			sge = &msg->sg.data[msg->sg.end];
			sg_unmark_end(sge);
			sg_set_page(sge, pfrag->page, use, orig_offset);
			get_page(pfrag->page);
			sk_msg_iter_next(msg, end);
		}

		sk_mem_charge(sk, use);
		msg->sg.size += use;
		pfrag->offset += use;
		len -= use;
	}

	return ret;
79 80 81 82

msg_trim:
	sk_msg_trim(sk, msg, osize);
	return ret;
83 84 85
}
EXPORT_SYMBOL_GPL(sk_msg_alloc);

86 87 88 89 90
int sk_msg_clone(struct sock *sk, struct sk_msg *dst, struct sk_msg *src,
		 u32 off, u32 len)
{
	int i = src->sg.start;
	struct scatterlist *sge = sk_msg_elem(src, i);
91
	struct scatterlist *sgd = NULL;
92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107
	u32 sge_len, sge_off;

	while (off) {
		if (sge->length > off)
			break;
		off -= sge->length;
		sk_msg_iter_var_next(i);
		if (i == src->sg.end && off)
			return -ENOSPC;
		sge = sk_msg_elem(src, i);
	}

	while (len) {
		sge_len = sge->length - off;
		if (sge_len > len)
			sge_len = len;
108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123

		if (dst->sg.end)
			sgd = sk_msg_elem(dst, dst->sg.end - 1);

		if (sgd &&
		    (sg_page(sge) == sg_page(sgd)) &&
		    (sg_virt(sge) + off == sg_virt(sgd) + sgd->length)) {
			sgd->length += sge_len;
			dst->sg.size += sge_len;
		} else if (!sk_msg_full(dst)) {
			sge_off = sge->offset + off;
			sk_msg_page_add(dst, sg_page(sge), sge_len, sge_off);
		} else {
			return -ENOSPC;
		}

124 125 126 127 128 129 130 131 132 133 134 135 136
		off = 0;
		len -= sge_len;
		sk_mem_charge(sk, sge_len);
		sk_msg_iter_var_next(i);
		if (i == src->sg.end && len)
			return -ENOSPC;
		sge = sk_msg_elem(src, i);
	}

	return 0;
}
EXPORT_SYMBOL_GPL(sk_msg_clone);

137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
void sk_msg_return_zero(struct sock *sk, struct sk_msg *msg, int bytes)
{
	int i = msg->sg.start;

	do {
		struct scatterlist *sge = sk_msg_elem(msg, i);

		if (bytes < sge->length) {
			sge->length -= bytes;
			sge->offset += bytes;
			sk_mem_uncharge(sk, bytes);
			break;
		}

		sk_mem_uncharge(sk, sge->length);
		bytes -= sge->length;
		sge->length = 0;
		sge->offset = 0;
		sk_msg_iter_var_next(i);
	} while (bytes && i != msg->sg.end);
	msg->sg.start = i;
}
EXPORT_SYMBOL_GPL(sk_msg_return_zero);

void sk_msg_return(struct sock *sk, struct sk_msg *msg, int bytes)
{
	int i = msg->sg.start;

	do {
		struct scatterlist *sge = &msg->sg.data[i];
		int uncharge = (bytes < sge->length) ? bytes : sge->length;

		sk_mem_uncharge(sk, uncharge);
		bytes -= uncharge;
		sk_msg_iter_var_next(i);
	} while (i != msg->sg.end);
}
EXPORT_SYMBOL_GPL(sk_msg_return);

static int sk_msg_free_elem(struct sock *sk, struct sk_msg *msg, u32 i,
			    bool charge)
{
	struct scatterlist *sge = sk_msg_elem(msg, i);
	u32 len = sge->length;

182 183 184 185
	/* When the skb owns the memory we free it from consume_skb path. */
	if (!msg->skb) {
		if (charge)
			sk_mem_uncharge(sk, len);
186
		put_page(sg_page(sge));
187
	}
188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204
	memset(sge, 0, sizeof(*sge));
	return len;
}

static int __sk_msg_free(struct sock *sk, struct sk_msg *msg, u32 i,
			 bool charge)
{
	struct scatterlist *sge = sk_msg_elem(msg, i);
	int freed = 0;

	while (msg->sg.size) {
		msg->sg.size -= sge->length;
		freed += sk_msg_free_elem(sk, msg, i, charge);
		sk_msg_iter_var_next(i);
		sk_msg_check_to_free(msg, i, msg->sg.size);
		sge = sk_msg_elem(msg, i);
	}
205
	consume_skb(msg->skb);
206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284
	sk_msg_init(msg);
	return freed;
}

int sk_msg_free_nocharge(struct sock *sk, struct sk_msg *msg)
{
	return __sk_msg_free(sk, msg, msg->sg.start, false);
}
EXPORT_SYMBOL_GPL(sk_msg_free_nocharge);

int sk_msg_free(struct sock *sk, struct sk_msg *msg)
{
	return __sk_msg_free(sk, msg, msg->sg.start, true);
}
EXPORT_SYMBOL_GPL(sk_msg_free);

static void __sk_msg_free_partial(struct sock *sk, struct sk_msg *msg,
				  u32 bytes, bool charge)
{
	struct scatterlist *sge;
	u32 i = msg->sg.start;

	while (bytes) {
		sge = sk_msg_elem(msg, i);
		if (!sge->length)
			break;
		if (bytes < sge->length) {
			if (charge)
				sk_mem_uncharge(sk, bytes);
			sge->length -= bytes;
			sge->offset += bytes;
			msg->sg.size -= bytes;
			break;
		}

		msg->sg.size -= sge->length;
		bytes -= sge->length;
		sk_msg_free_elem(sk, msg, i, charge);
		sk_msg_iter_var_next(i);
		sk_msg_check_to_free(msg, i, bytes);
	}
	msg->sg.start = i;
}

void sk_msg_free_partial(struct sock *sk, struct sk_msg *msg, u32 bytes)
{
	__sk_msg_free_partial(sk, msg, bytes, true);
}
EXPORT_SYMBOL_GPL(sk_msg_free_partial);

void sk_msg_free_partial_nocharge(struct sock *sk, struct sk_msg *msg,
				  u32 bytes)
{
	__sk_msg_free_partial(sk, msg, bytes, false);
}

void sk_msg_trim(struct sock *sk, struct sk_msg *msg, int len)
{
	int trim = msg->sg.size - len;
	u32 i = msg->sg.end;

	if (trim <= 0) {
		WARN_ON(trim < 0);
		return;
	}

	sk_msg_iter_var_prev(i);
	msg->sg.size = len;
	while (msg->sg.data[i].length &&
	       trim >= msg->sg.data[i].length) {
		trim -= msg->sg.data[i].length;
		sk_msg_free_elem(sk, msg, i, true);
		sk_msg_iter_var_prev(i);
		if (!trim)
			goto out;
	}

	msg->sg.data[i].length -= trim;
	sk_mem_uncharge(sk, trim);
285 286 287
	/* Adjust copybreak if it falls into the trimmed part of last buf */
	if (msg->sg.curr == i && msg->sg.copybreak > msg->sg.data[i].length)
		msg->sg.copybreak = msg->sg.data[i].length;
288
out:
289 290 291 292 293 294
	sk_msg_iter_var_next(i);
	msg->sg.end = i;

	/* If we trim data a full sg elem before curr pointer update
	 * copybreak and current so that any future copy operations
	 * start at new copy location.
295 296 297
	 * However trimed data that has not yet been used in a copy op
	 * does not require an update.
	 */
298 299 300 301 302 303
	if (!msg->sg.size) {
		msg->sg.curr = msg->sg.start;
		msg->sg.copybreak = 0;
	} else if (sk_msg_iter_dist(msg->sg.start, msg->sg.curr) >=
		   sk_msg_iter_dist(msg->sg.start, msg->sg.end)) {
		sk_msg_iter_var_prev(i);
304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326
		msg->sg.curr = i;
		msg->sg.copybreak = msg->sg.data[i].length;
	}
}
EXPORT_SYMBOL_GPL(sk_msg_trim);

int sk_msg_zerocopy_from_iter(struct sock *sk, struct iov_iter *from,
			      struct sk_msg *msg, u32 bytes)
{
	int i, maxpages, ret = 0, num_elems = sk_msg_elem_used(msg);
	const int to_max_pages = MAX_MSG_FRAGS;
	struct page *pages[MAX_MSG_FRAGS];
	ssize_t orig, copied, use, offset;

	orig = msg->sg.size;
	while (bytes > 0) {
		i = 0;
		maxpages = to_max_pages - num_elems;
		if (maxpages == 0) {
			ret = -EFAULT;
			goto out;
		}

327
		copied = iov_iter_get_pages2(from, pages, bytes, maxpages,
328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409
					    &offset);
		if (copied <= 0) {
			ret = -EFAULT;
			goto out;
		}

		bytes -= copied;
		msg->sg.size += copied;

		while (copied) {
			use = min_t(int, copied, PAGE_SIZE - offset);
			sg_set_page(&msg->sg.data[msg->sg.end],
				    pages[i], use, offset);
			sg_unmark_end(&msg->sg.data[msg->sg.end]);
			sk_mem_charge(sk, use);

			offset = 0;
			copied -= use;
			sk_msg_iter_next(msg, end);
			num_elems++;
			i++;
		}
		/* When zerocopy is mixed with sk_msg_*copy* operations we
		 * may have a copybreak set in this case clear and prefer
		 * zerocopy remainder when possible.
		 */
		msg->sg.copybreak = 0;
		msg->sg.curr = msg->sg.end;
	}
out:
	/* Revert iov_iter updates, msg will need to use 'trim' later if it
	 * also needs to be cleared.
	 */
	if (ret)
		iov_iter_revert(from, msg->sg.size - orig);
	return ret;
}
EXPORT_SYMBOL_GPL(sk_msg_zerocopy_from_iter);

int sk_msg_memcopy_from_iter(struct sock *sk, struct iov_iter *from,
			     struct sk_msg *msg, u32 bytes)
{
	int ret = -ENOSPC, i = msg->sg.curr;
	struct scatterlist *sge;
	u32 copy, buf_size;
	void *to;

	do {
		sge = sk_msg_elem(msg, i);
		/* This is possible if a trim operation shrunk the buffer */
		if (msg->sg.copybreak >= sge->length) {
			msg->sg.copybreak = 0;
			sk_msg_iter_var_next(i);
			if (i == msg->sg.end)
				break;
			sge = sk_msg_elem(msg, i);
		}

		buf_size = sge->length - msg->sg.copybreak;
		copy = (buf_size > bytes) ? bytes : buf_size;
		to = sg_virt(sge) + msg->sg.copybreak;
		msg->sg.copybreak += copy;
		if (sk->sk_route_caps & NETIF_F_NOCACHE_COPY)
			ret = copy_from_iter_nocache(to, copy, from);
		else
			ret = copy_from_iter(to, copy, from);
		if (ret != copy) {
			ret = -EFAULT;
			goto out;
		}
		bytes -= copy;
		if (!bytes)
			break;
		msg->sg.copybreak = 0;
		sk_msg_iter_var_next(i);
	} while (i != msg->sg.end);
out:
	msg->sg.curr = i;
	return ret;
}
EXPORT_SYMBOL_GPL(sk_msg_memcopy_from_iter);

410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
/* Receive sk_msg from psock->ingress_msg to @msg. */
int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
		   int len, int flags)
{
	struct iov_iter *iter = &msg->msg_iter;
	int peek = flags & MSG_PEEK;
	struct sk_msg *msg_rx;
	int i, copied = 0;

	msg_rx = sk_psock_peek_msg(psock);
	while (copied != len) {
		struct scatterlist *sge;

		if (unlikely(!msg_rx))
			break;

		i = msg_rx->sg.start;
		do {
			struct page *page;
			int copy;

			sge = sk_msg_elem(msg_rx, i);
			copy = sge->length;
			page = sg_page(sge);
			if (copied + copy > len)
				copy = len - copied;
			copy = copy_page_to_iter(page, sge->offset, copy, iter);
			if (!copy)
				return copied ? copied : -EFAULT;

			copied += copy;
			if (likely(!peek)) {
				sge->offset += copy;
				sge->length -= copy;
				if (!msg_rx->skb)
					sk_mem_uncharge(sk, copy);
				msg_rx->sg.size -= copy;

				if (!sge->length) {
					sk_msg_iter_var_next(i);
					if (!msg_rx->skb)
						put_page(page);
				}
			} else {
				/* Lets not optimize peek case if copy_page_to_iter
				 * didn't copy the entire length lets just break.
				 */
				if (copy != sge->length)
					return copied;
				sk_msg_iter_var_next(i);
			}

			if (copied == len)
				break;
464
		} while (!sg_is_last(sge));
465 466 467 468 469 470 471 472 473

		if (unlikely(peek)) {
			msg_rx = sk_psock_next_msg(psock, msg_rx);
			if (!msg_rx)
				break;
			continue;
		}

		msg_rx->sg.start = i;
474
		if (!sge->length && sg_is_last(sge)) {
475 476 477 478 479 480 481 482 483 484
			msg_rx = sk_psock_dequeue_msg(psock);
			kfree_sk_msg(msg_rx);
		}
		msg_rx = sk_psock_peek_msg(psock);
	}

	return copied;
}
EXPORT_SYMBOL_GPL(sk_msg_recvmsg);

485 486 487 488 489 490 491 492 493 494 495 496 497 498
bool sk_msg_is_readable(struct sock *sk)
{
	struct sk_psock *psock;
	bool empty = true;

	rcu_read_lock();
	psock = sk_psock(sk);
	if (likely(psock))
		empty = list_empty(&psock->ingress_msg);
	rcu_read_unlock();
	return !empty;
}
EXPORT_SYMBOL_GPL(sk_msg_is_readable);

499
static struct sk_msg *alloc_sk_msg(void)
500 501 502
{
	struct sk_msg *msg;

503 504
	msg = kzalloc(sizeof(*msg), __GFP_NOWARN | GFP_KERNEL);
	if (unlikely(!msg))
505
		return NULL;
506 507 508
	sg_init_marker(msg->sg.data, NR_MSG_FRAG_IDS);
	return msg;
}
509

510 511 512 513
static struct sk_msg *sk_psock_create_ingress_msg(struct sock *sk,
						  struct sk_buff *skb)
{
	if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf)
514
		return NULL;
515

516
	if (!sk_rmem_schedule(sk, skb, skb->truesize))
517
		return NULL;
518

519
	return alloc_sk_msg();
520 521 522
}

static int sk_psock_skb_ingress_enqueue(struct sk_buff *skb,
523
					u32 off, u32 len,
524 525 526 527
					struct sk_psock *psock,
					struct sock *sk,
					struct sk_msg *msg)
{
528
	int num_sge, copied;
529

530
	num_sge = skb_to_sgvec(skb, msg->sg.data, off, len);
531 532 533 534 535 536 537 538 539 540 541 542 543
	if (num_sge < 0) {
		/* skb linearize may fail with ENOMEM, but lets simply try again
		 * later if this happens. Under memory pressure we don't want to
		 * drop the skb. We need to linearize the skb so that the mapping
		 * in skb_to_sgvec can not error.
		 */
		if (skb_linearize(skb))
			return -EAGAIN;

		num_sge = skb_to_sgvec(skb, msg->sg.data, off, len);
		if (unlikely(num_sge < 0))
			return num_sge;
	}
544

545
	copied = len;
546
	msg->sg.start = 0;
547
	msg->sg.size = copied;
548
	msg->sg.end = num_sge;
549 550 551
	msg->skb = skb;

	sk_psock_queue_msg(psock, msg);
552
	sk_psock_data_ready(sk, psock);
553 554 555
	return copied;
}

556 557
static int sk_psock_skb_ingress_self(struct sk_psock *psock, struct sk_buff *skb,
				     u32 off, u32 len);
558

559 560
static int sk_psock_skb_ingress(struct sk_psock *psock, struct sk_buff *skb,
				u32 off, u32 len)
561 562 563
{
	struct sock *sk = psock->sk;
	struct sk_msg *msg;
564
	int err;
565

566 567 568 569 570
	/* If we are receiving on the same sock skb->sk is already assigned,
	 * skip memory accounting and owner transition seeing it already set
	 * correctly.
	 */
	if (unlikely(skb->sk == sk))
571
		return sk_psock_skb_ingress_self(psock, skb, off, len);
572 573 574 575 576 577 578 579 580 581 582
	msg = sk_psock_create_ingress_msg(sk, skb);
	if (!msg)
		return -EAGAIN;

	/* This will transition ownership of the data from the socket where
	 * the BPF program was run initiating the redirect to the socket
	 * we will eventually receive this data on. The data will be released
	 * from skb_consume found in __tcp_bpf_recvmsg() after its been copied
	 * into user buffers.
	 */
	skb_set_owner_r(skb, sk);
583
	err = sk_psock_skb_ingress_enqueue(skb, off, len, psock, sk, msg);
584 585 586
	if (err < 0)
		kfree(msg);
	return err;
587 588 589 590 591 592
}

/* Puts an skb on the ingress queue of the socket already assigned to the
 * skb. In this case we do not need to check memory limits or skb_set_owner_r
 * because the skb is already accounted for here.
 */
593 594
static int sk_psock_skb_ingress_self(struct sk_psock *psock, struct sk_buff *skb,
				     u32 off, u32 len)
595
{
596
	struct sk_msg *msg = alloc_sk_msg();
597
	struct sock *sk = psock->sk;
598
	int err;
599 600 601

	if (unlikely(!msg))
		return -EAGAIN;
602
	skb_set_owner_r(skb, sk);
603
	err = sk_psock_skb_ingress_enqueue(skb, off, len, psock, sk, msg);
604 605 606
	if (err < 0)
		kfree(msg);
	return err;
607 608
}

609 610 611
static int sk_psock_handle_skb(struct sk_psock *psock, struct sk_buff *skb,
			       u32 off, u32 len, bool ingress)
{
612 613 614
	if (!ingress) {
		if (!sock_writeable(psock->sk))
			return -EAGAIN;
615
		return skb_send_sock(psock->sk, skb, off, len);
616
	}
617
	return sk_psock_skb_ingress(psock, skb, off, len);
618 619
}

620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635
static void sk_psock_skb_state(struct sk_psock *psock,
			       struct sk_psock_work_state *state,
			       struct sk_buff *skb,
			       int len, int off)
{
	spin_lock_bh(&psock->ingress_lock);
	if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) {
		state->skb = skb;
		state->len = len;
		state->off = off;
	} else {
		sock_drop(psock->sk, skb);
	}
	spin_unlock_bh(&psock->ingress_lock);
}

636 637 638 639
static void sk_psock_backlog(struct work_struct *work)
{
	struct sk_psock *psock = container_of(work, struct sk_psock, work);
	struct sk_psock_work_state *state = &psock->work_state;
640
	struct sk_buff *skb = NULL;
641 642 643 644
	bool ingress;
	u32 len, off;
	int ret;

645
	mutex_lock(&psock->work_mutex);
646 647
	if (unlikely(state->skb)) {
		spin_lock_bh(&psock->ingress_lock);
648 649 650 651
		skb = state->skb;
		len = state->len;
		off = state->off;
		state->skb = NULL;
652
		spin_unlock_bh(&psock->ingress_lock);
653
	}
654 655
	if (skb)
		goto start;
656 657 658 659

	while ((skb = skb_dequeue(&psock->ingress_skb))) {
		len = skb->len;
		off = 0;
660 661 662 663 664 665
		if (skb_bpf_strparser(skb)) {
			struct strp_msg *stm = strp_msg(skb);

			off = stm->offset;
			len = stm->full_len;
		}
666
start:
667 668
		ingress = skb_bpf_ingress(skb);
		skb_bpf_redirect_clear(skb);
669 670
		do {
			ret = -EIO;
671
			if (!sock_flag(psock->sk, SOCK_DEAD))
672 673 674 675
				ret = sk_psock_handle_skb(psock, skb, off,
							  len, ingress);
			if (ret <= 0) {
				if (ret == -EAGAIN) {
676 677
					sk_psock_skb_state(psock, state, skb,
							   len, off);
678 679 680 681 682
					goto end;
				}
				/* Hard errors break pipe and stop xmit. */
				sk_psock_report_error(psock, ret ? -ret : EPIPE);
				sk_psock_clear_state(psock, SK_PSOCK_TX_ENABLED);
683
				sock_drop(psock->sk, skb);
684 685 686 687 688 689 690 691 692 693
				goto end;
			}
			off += ret;
			len -= ret;
		} while (len);

		if (!ingress)
			kfree_skb(skb);
	}
end:
694
	mutex_unlock(&psock->work_mutex);
695 696 697 698
}

struct sk_psock *sk_psock_init(struct sock *sk, int node)
{
699 700
	struct sk_psock *psock;
	struct proto *prot;
701

702 703
	write_lock_bh(&sk->sk_callback_lock);

704 705 706 707 708
	if (sk_is_inet(sk) && inet_csk_has_ulp(sk)) {
		psock = ERR_PTR(-EINVAL);
		goto out;
	}

709 710 711 712 713 714 715 716 717 718 719 720
	if (sk->sk_user_data) {
		psock = ERR_PTR(-EBUSY);
		goto out;
	}

	psock = kzalloc_node(sizeof(*psock), GFP_ATOMIC | __GFP_NOWARN, node);
	if (!psock) {
		psock = ERR_PTR(-ENOMEM);
		goto out;
	}

	prot = READ_ONCE(sk->sk_prot);
721
	psock->sk = sk;
722 723 724
	psock->eval = __SK_NONE;
	psock->sk_proto = prot;
	psock->saved_unhash = prot->unhash;
725
	psock->saved_destroy = prot->destroy;
726 727
	psock->saved_close = prot->close;
	psock->saved_write_space = sk->sk_write_space;
728 729 730 731 732

	INIT_LIST_HEAD(&psock->link);
	spin_lock_init(&psock->link_lock);

	INIT_WORK(&psock->work, sk_psock_backlog);
733
	mutex_init(&psock->work_mutex);
734
	INIT_LIST_HEAD(&psock->ingress_msg);
735
	spin_lock_init(&psock->ingress_lock);
736 737 738 739 740
	skb_queue_head_init(&psock->ingress_skb);

	sk_psock_set_state(psock, SK_PSOCK_TX_ENABLED);
	refcount_set(&psock->refcnt, 1);

741
	rcu_assign_sk_user_data_nocopy(sk, psock);
742 743
	sock_hold(sk);

744 745
out:
	write_unlock_bh(&sk->sk_callback_lock);
746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762
	return psock;
}
EXPORT_SYMBOL_GPL(sk_psock_init);

struct sk_psock_link *sk_psock_link_pop(struct sk_psock *psock)
{
	struct sk_psock_link *link;

	spin_lock_bh(&psock->link_lock);
	link = list_first_entry_or_null(&psock->link, struct sk_psock_link,
					list);
	if (link)
		list_del(&link->list);
	spin_unlock_bh(&psock->link_lock);
	return link;
}

763
static void __sk_psock_purge_ingress_msg(struct sk_psock *psock)
764 765 766 767 768 769 770 771 772 773
{
	struct sk_msg *msg, *tmp;

	list_for_each_entry_safe(msg, tmp, &psock->ingress_msg, list) {
		list_del(&msg->list);
		sk_msg_free(psock->sk, msg);
		kfree(msg);
	}
}

774
static void __sk_psock_zap_ingress(struct sk_psock *psock)
775
{
776 777
	struct sk_buff *skb;

778
	while ((skb = skb_dequeue(&psock->ingress_skb)) != NULL) {
779
		skb_bpf_redirect_clear(skb);
780
		sock_drop(psock->sk, skb);
781
	}
782 783 784 785 786
	kfree_skb(psock->work_state.skb);
	/* We null the skb here to ensure that calls to sk_psock_backlog
	 * do not pick up the free'd skb.
	 */
	psock->work_state.skb = NULL;
787 788 789 790 791 792 793 794 795 796 797 798 799
	__sk_psock_purge_ingress_msg(psock);
}

static void sk_psock_link_destroy(struct sk_psock *psock)
{
	struct sk_psock_link *link, *tmp;

	list_for_each_entry_safe(link, tmp, &psock->link, list) {
		list_del(&link->list);
		sk_psock_free_link(link);
	}
}

800 801 802 803 804 805 806 807 808 809 810 811
void sk_psock_stop(struct sk_psock *psock, bool wait)
{
	spin_lock_bh(&psock->ingress_lock);
	sk_psock_clear_state(psock, SK_PSOCK_TX_ENABLED);
	sk_psock_cork_free(psock);
	__sk_psock_zap_ingress(psock);
	spin_unlock_bh(&psock->ingress_lock);

	if (wait)
		cancel_work_sync(&psock->work);
}

812 813
static void sk_psock_done_strp(struct sk_psock *psock);

814
static void sk_psock_destroy(struct work_struct *work)
815
{
816 817
	struct sk_psock *psock = container_of(to_rcu_work(work),
					      struct sk_psock, rwork);
818
	/* No sk_callback_lock since already detached. */
819

820
	sk_psock_done_strp(psock);
821 822

	cancel_work_sync(&psock->work);
823
	mutex_destroy(&psock->work_mutex);
824 825 826 827 828 829 830 831 832 833 834 835 836 837 838

	psock_progs_drop(&psock->progs);

	sk_psock_link_destroy(psock);
	sk_psock_cork_free(psock);

	if (psock->sk_redir)
		sock_put(psock->sk_redir);
	sock_put(psock->sk);
	kfree(psock);
}

void sk_psock_drop(struct sock *sk, struct sk_psock *psock)
{
	write_lock_bh(&sk->sk_callback_lock);
839 840
	sk_psock_restore_proto(sk, psock);
	rcu_assign_sk_user_data(sk, NULL);
841
	if (psock->progs.stream_parser)
842
		sk_psock_stop_strp(sk, psock);
843
	else if (psock->progs.stream_verdict || psock->progs.skb_verdict)
844
		sk_psock_stop_verdict(sk, psock);
845 846
	write_unlock_bh(&sk->sk_callback_lock);

847 848
	sk_psock_stop(psock, false);

849 850
	INIT_RCU_WORK(&psock->rwork, sk_psock_destroy);
	queue_rcu_work(system_wq, &psock->rwork);
851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881
}
EXPORT_SYMBOL_GPL(sk_psock_drop);

static int sk_psock_map_verd(int verdict, bool redir)
{
	switch (verdict) {
	case SK_PASS:
		return redir ? __SK_REDIRECT : __SK_PASS;
	case SK_DROP:
	default:
		break;
	}

	return __SK_DROP;
}

int sk_psock_msg_verdict(struct sock *sk, struct sk_psock *psock,
			 struct sk_msg *msg)
{
	struct bpf_prog *prog;
	int ret;

	rcu_read_lock();
	prog = READ_ONCE(psock->progs.msg_parser);
	if (unlikely(!prog)) {
		ret = __SK_PASS;
		goto out;
	}

	sk_msg_compute_data_pointers(msg);
	msg->sk = sk;
882
	ret = bpf_prog_run_pin_on_cpu(prog, msg);
883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900
	ret = sk_psock_map_verd(ret, msg->sk_redir);
	psock->apply_bytes = msg->apply_bytes;
	if (ret == __SK_REDIRECT) {
		if (psock->sk_redir)
			sock_put(psock->sk_redir);
		psock->sk_redir = msg->sk_redir;
		if (!psock->sk_redir) {
			ret = __SK_DROP;
			goto out;
		}
		sock_hold(psock->sk_redir);
	}
out:
	rcu_read_unlock();
	return ret;
}
EXPORT_SYMBOL_GPL(sk_psock_msg_verdict);

901
static int sk_psock_skb_redirect(struct sk_psock *from, struct sk_buff *skb)
902 903 904 905
{
	struct sk_psock *psock_other;
	struct sock *sk_other;

906
	sk_other = skb_bpf_redirect_fetch(skb);
907 908 909
	/* This error is a buggy BPF program, it returned a redirect
	 * return code, but then didn't set a redirect interface.
	 */
910
	if (unlikely(!sk_other)) {
911
		skb_bpf_redirect_clear(skb);
912
		sock_drop(from->sk, skb);
913
		return -EIO;
914 915
	}
	psock_other = sk_psock(sk_other);
916 917 918 919
	/* This error indicates the socket is being torn down or had another
	 * error that caused the pipe to break. We can't send a packet on
	 * a socket that is in this state so we drop the skb.
	 */
920
	if (!psock_other || sock_flag(sk_other, SOCK_DEAD)) {
921
		skb_bpf_redirect_clear(skb);
922
		sock_drop(from->sk, skb);
923
		return -EIO;
924 925 926 927
	}
	spin_lock_bh(&psock_other->ingress_lock);
	if (!sk_psock_test_state(psock_other, SK_PSOCK_TX_ENABLED)) {
		spin_unlock_bh(&psock_other->ingress_lock);
928
		skb_bpf_redirect_clear(skb);
929
		sock_drop(from->sk, skb);
930
		return -EIO;
931 932
	}

933 934
	skb_queue_tail(&psock_other->ingress_skb, skb);
	schedule_work(&psock_other->work);
935
	spin_unlock_bh(&psock_other->ingress_lock);
936
	return 0;
937 938
}

939 940
static void sk_psock_tls_verdict_apply(struct sk_buff *skb,
				       struct sk_psock *from, int verdict)
941 942 943
{
	switch (verdict) {
	case __SK_REDIRECT:
944
		sk_psock_skb_redirect(from, skb);
945 946 947 948 949 950 951 952 953 954 955 956 957 958
		break;
	case __SK_PASS:
	case __SK_DROP:
	default:
		break;
	}
}

int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb)
{
	struct bpf_prog *prog;
	int ret = __SK_PASS;

	rcu_read_lock();
959
	prog = READ_ONCE(psock->progs.stream_verdict);
960
	if (likely(prog)) {
961
		skb->sk = psock->sk;
962 963
		skb_dst_drop(skb);
		skb_bpf_redirect_clear(skb);
964
		ret = bpf_prog_run_pin_on_cpu(prog, skb);
965
		ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
966
		skb->sk = NULL;
967
	}
968
	sk_psock_tls_verdict_apply(skb, psock, ret);
969 970 971 972 973
	rcu_read_unlock();
	return ret;
}
EXPORT_SYMBOL_GPL(sk_psock_tls_strp_read);

974 975
static int sk_psock_verdict_apply(struct sk_psock *psock, struct sk_buff *skb,
				  int verdict)
976 977
{
	struct sock *sk_other;
978
	int err = 0;
979
	u32 len, off;
980

981
	switch (verdict) {
982
	case __SK_PASS:
983
		err = -EIO;
984 985 986
		sk_other = psock->sk;
		if (sock_flag(sk_other, SOCK_DEAD) ||
		    !sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) {
987
			skb_bpf_redirect_clear(skb);
988 989 990
			goto out_free;
		}

991
		skb_bpf_set_ingress(skb);
992 993 994 995 996 997 998 999

		/* If the queue is empty then we can submit directly
		 * into the msg queue. If its not empty we have to
		 * queue work otherwise we may get OOO data. Otherwise,
		 * if sk_psock_skb_ingress errors will be handled by
		 * retrying later from workqueue.
		 */
		if (skb_queue_empty(&psock->ingress_skb)) {
1000 1001 1002 1003 1004 1005 1006 1007 1008
			len = skb->len;
			off = 0;
			if (skb_bpf_strparser(skb)) {
				struct strp_msg *stm = strp_msg(skb);

				off = stm->offset;
				len = stm->full_len;
			}
			err = sk_psock_skb_ingress_self(psock, skb, off, len);
1009 1010
		}
		if (err < 0) {
1011 1012 1013 1014
			spin_lock_bh(&psock->ingress_lock);
			if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) {
				skb_queue_tail(&psock->ingress_skb, skb);
				schedule_work(&psock->work);
1015
				err = 0;
1016 1017
			}
			spin_unlock_bh(&psock->ingress_lock);
1018 1019 1020 1021
			if (err < 0) {
				skb_bpf_redirect_clear(skb);
				goto out_free;
			}
1022
		}
1023
		break;
1024
	case __SK_REDIRECT:
1025
		err = sk_psock_skb_redirect(psock, skb);
1026
		break;
1027 1028 1029
	case __SK_DROP:
	default:
out_free:
1030
		sock_drop(psock->sk, skb);
1031
	}
1032 1033

	return err;
1034 1035
}

1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053
static void sk_psock_write_space(struct sock *sk)
{
	struct sk_psock *psock;
	void (*write_space)(struct sock *sk) = NULL;

	rcu_read_lock();
	psock = sk_psock(sk);
	if (likely(psock)) {
		if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED))
			schedule_work(&psock->work);
		write_space = psock->saved_write_space;
	}
	rcu_read_unlock();
	if (write_space)
		write_space(sk);
}

#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
1054 1055
static void sk_psock_strp_read(struct strparser *strp, struct sk_buff *skb)
{
1056
	struct sk_psock *psock;
1057 1058
	struct bpf_prog *prog;
	int ret = __SK_DROP;
1059
	struct sock *sk;
1060 1061

	rcu_read_lock();
1062 1063 1064
	sk = strp->sk;
	psock = sk_psock(sk);
	if (unlikely(!psock)) {
1065
		sock_drop(sk, skb);
1066 1067
		goto out;
	}
1068
	prog = READ_ONCE(psock->progs.stream_verdict);
1069
	if (likely(prog)) {
1070
		skb->sk = sk;
1071 1072
		skb_dst_drop(skb);
		skb_bpf_redirect_clear(skb);
1073
		ret = bpf_prog_run_pin_on_cpu(prog, skb);
1074 1075
		if (ret == SK_PASS)
			skb_bpf_set_strparser(skb);
1076
		ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
1077
		skb->sk = NULL;
1078 1079
	}
	sk_psock_verdict_apply(psock, skb, ret);
1080
out:
1081
	rcu_read_unlock();
1082 1083 1084 1085 1086 1087 1088 1089 1090
}

static int sk_psock_strp_read_done(struct strparser *strp, int err)
{
	return err;
}

static int sk_psock_strp_parse(struct strparser *strp, struct sk_buff *skb)
{
1091
	struct sk_psock *psock = container_of(strp, struct sk_psock, strp);
1092 1093 1094 1095
	struct bpf_prog *prog;
	int ret = skb->len;

	rcu_read_lock();
1096
	prog = READ_ONCE(psock->progs.stream_parser);
1097 1098
	if (likely(prog)) {
		skb->sk = psock->sk;
1099
		ret = bpf_prog_run_pin_on_cpu(prog, skb);
1100 1101
		skb->sk = NULL;
	}
1102 1103 1104 1105 1106
	rcu_read_unlock();
	return ret;
}

/* Called with socket lock held. */
1107
static void sk_psock_strp_data_ready(struct sock *sk)
1108 1109 1110 1111 1112 1113
{
	struct sk_psock *psock;

	rcu_read_lock();
	psock = sk_psock(sk);
	if (likely(psock)) {
1114
		if (tls_sw_has_ctx_rx(sk)) {
1115
			psock->saved_data_ready(sk);
1116 1117
		} else {
			write_lock_bh(&sk->sk_callback_lock);
1118
			strp_data_ready(&psock->strp);
1119 1120
			write_unlock_bh(&sk->sk_callback_lock);
		}
1121 1122 1123 1124
	}
	rcu_read_unlock();
}

1125 1126 1127 1128 1129 1130 1131 1132
int sk_psock_init_strp(struct sock *sk, struct sk_psock *psock)
{
	static const struct strp_callbacks cb = {
		.rcv_msg	= sk_psock_strp_read,
		.read_sock_done	= sk_psock_strp_read_done,
		.parse_msg	= sk_psock_strp_parse,
	};

1133
	return strp_init(&psock->strp, sk, &cb);
1134 1135 1136 1137
}

void sk_psock_start_strp(struct sock *sk, struct sk_psock *psock)
{
1138
	if (psock->saved_data_ready)
1139 1140
		return;

1141
	psock->saved_data_ready = sk->sk_data_ready;
1142 1143 1144 1145 1146 1147
	sk->sk_data_ready = sk_psock_strp_data_ready;
	sk->sk_write_space = sk_psock_write_space;
}

void sk_psock_stop_strp(struct sock *sk, struct sk_psock *psock)
{
1148 1149
	psock_set_prog(&psock->progs.stream_parser, NULL);

1150
	if (!psock->saved_data_ready)
1151 1152
		return;

1153 1154 1155
	sk->sk_data_ready = psock->saved_data_ready;
	psock->saved_data_ready = NULL;
	strp_stop(&psock->strp);
1156 1157 1158 1159 1160
}

static void sk_psock_done_strp(struct sk_psock *psock)
{
	/* Parser has been stopped */
1161
	if (psock->progs.stream_parser)
1162
		strp_done(&psock->strp);
1163 1164 1165 1166 1167 1168 1169
}
#else
static void sk_psock_done_strp(struct sk_psock *psock)
{
}
#endif /* CONFIG_BPF_STREAM_PARSER */

1170
static int sk_psock_verdict_recv(struct sock *sk, struct sk_buff *skb)
1171 1172 1173 1174
{
	struct sk_psock *psock;
	struct bpf_prog *prog;
	int ret = __SK_DROP;
1175
	int len = skb->len;
1176

1177
	skb_get(skb);
1178 1179 1180 1181 1182

	rcu_read_lock();
	psock = sk_psock(sk);
	if (unlikely(!psock)) {
		len = 0;
1183
		sock_drop(sk, skb);
1184 1185
		goto out;
	}
1186
	prog = READ_ONCE(psock->progs.stream_verdict);
1187 1188
	if (!prog)
		prog = READ_ONCE(psock->progs.skb_verdict);
1189
	if (likely(prog)) {
1190 1191
		skb_dst_drop(skb);
		skb_bpf_redirect_clear(skb);
1192
		ret = bpf_prog_run_pin_on_cpu(prog, skb);
1193
		ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb));
1194
	}
1195 1196
	if (sk_psock_verdict_apply(psock, skb, ret) < 0)
		len = 0;
1197 1198 1199 1200 1201 1202 1203 1204 1205
out:
	rcu_read_unlock();
	return len;
}

static void sk_psock_verdict_data_ready(struct sock *sk)
{
	struct socket *sock = sk->sk_socket;

1206
	if (unlikely(!sock || !sock->ops || !sock->ops->read_skb))
1207
		return;
1208
	sock->ops->read_skb(sk, sk_psock_verdict_recv);
1209 1210 1211 1212
}

void sk_psock_start_verdict(struct sock *sk, struct sk_psock *psock)
{
1213
	if (psock->saved_data_ready)
1214 1215
		return;

1216
	psock->saved_data_ready = sk->sk_data_ready;
1217 1218 1219 1220 1221 1222
	sk->sk_data_ready = sk_psock_verdict_data_ready;
	sk->sk_write_space = sk_psock_write_space;
}

void sk_psock_stop_verdict(struct sock *sk, struct sk_psock *psock)
{
1223 1224 1225
	psock_set_prog(&psock->progs.stream_verdict, NULL);
	psock_set_prog(&psock->progs.skb_verdict, NULL);

1226
	if (!psock->saved_data_ready)
1227 1228
		return;

1229 1230
	sk->sk_data_ready = psock->saved_data_ready;
	psock->saved_data_ready = NULL;
1231
}
新手
引导
客服 返回
顶部