af_packet.c 97.7 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5 6 7
/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		PACKET - implements raw packet sockets.
 *
8
 * Authors:	Ross Biro
Linus Torvalds's avatar
Linus Torvalds committed
9 10 11
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Alan Cox, <gw4pts@gw4pts.ampr.org>
 *
12
 * Fixes:
Linus Torvalds's avatar
Linus Torvalds committed
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
 *		Alan Cox	:	verify_area() now used correctly
 *		Alan Cox	:	new skbuff lists, look ma no backlogs!
 *		Alan Cox	:	tidied skbuff lists.
 *		Alan Cox	:	Now uses generic datagram routines I
 *					added. Also fixed the peek/read crash
 *					from all old Linux datagram code.
 *		Alan Cox	:	Uses the improved datagram code.
 *		Alan Cox	:	Added NULL's for socket options.
 *		Alan Cox	:	Re-commented the code.
 *		Alan Cox	:	Use new kernel side addressing
 *		Rob Janssen	:	Correct MTU usage.
 *		Dave Platt	:	Counter leaks caused by incorrect
 *					interrupt locking and some slightly
 *					dubious gcc output. Can you read
 *					compiler: it said _VOLATILE_
 *	Richard Kooijman	:	Timestamp fixes.
 *		Alan Cox	:	New buffers. Use sk->mac.raw.
 *		Alan Cox	:	sendmsg/recvmsg support.
 *		Alan Cox	:	Protocol setting support
 *	Alexey Kuznetsov	:	Untied from IPv4 stack.
 *	Cyrus Durgin		:	Fixed kerneld for kmod.
 *	Michal Ostrowski        :       Module initialization cleanup.
35
 *         Ulises Alonso        :       Frame number limit removal and
Linus Torvalds's avatar
Linus Torvalds committed
36
 *                                      packet_set_ring memory leak.
37 38 39
 *		Eric Biederman	:	Allow for > 8 byte hardware addresses.
 *					The convention is that longer addresses
 *					will simply extend the hardware address
40
 *					byte arrays at the end of sockaddr_ll
41
 *					and packet_mreq.
Johann Baudy's avatar
Johann Baudy committed
42
 *		Johann Baudy	:	Added TX RING.
43 44 45 46
 *		Chetan Loke	:	Implemented TPACKET_V3 block abstraction
 *					layer.
 *					Copyright (C) 2011, <lokec@ccs.neu.edu>
 *
Linus Torvalds's avatar
Linus Torvalds committed
47 48 49 50 51 52 53
 *
 *		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.
 *
 */
54

Linus Torvalds's avatar
Linus Torvalds committed
55 56
#include <linux/types.h>
#include <linux/mm.h>
57
#include <linux/capability.h>
Linus Torvalds's avatar
Linus Torvalds committed
58 59 60 61 62 63 64
#include <linux/fcntl.h>
#include <linux/socket.h>
#include <linux/in.h>
#include <linux/inet.h>
#include <linux/netdevice.h>
#include <linux/if_packet.h>
#include <linux/wireless.h>
65
#include <linux/kernel.h>
Linus Torvalds's avatar
Linus Torvalds committed
66
#include <linux/kmod.h>
67
#include <linux/slab.h>
68
#include <linux/vmalloc.h>
69
#include <net/net_namespace.h>
Linus Torvalds's avatar
Linus Torvalds committed
70 71 72 73 74 75 76 77 78
#include <net/ip.h>
#include <net/protocol.h>
#include <linux/skbuff.h>
#include <net/sock.h>
#include <linux/errno.h>
#include <linux/timer.h>
#include <asm/uaccess.h>
#include <asm/ioctls.h>
#include <asm/page.h>
79
#include <asm/cacheflush.h>
Linus Torvalds's avatar
Linus Torvalds committed
80 81 82 83 84 85
#include <asm/io.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/poll.h>
#include <linux/module.h>
#include <linux/init.h>
86
#include <linux/mutex.h>
87
#include <linux/if_vlan.h>
88
#include <linux/virtio_net.h>
89
#include <linux/errqueue.h>
90
#include <linux/net_tstamp.h>
91
#include <linux/percpu.h>
Linus Torvalds's avatar
Linus Torvalds committed
92 93 94 95
#ifdef CONFIG_INET
#include <net/inet_common.h>
#endif

96 97
#include "internal.h"

Linus Torvalds's avatar
Linus Torvalds committed
98 99 100 101 102 103 104 105 106 107 108 109 110 111 112
/*
   Assumptions:
   - if device has no dev->hard_header routine, it adds and removes ll header
     inside itself. In this case ll header is invisible outside of device,
     but higher levels still should reserve dev->hard_header_len.
     Some devices are enough clever to reallocate skb, when header
     will not fit to reserved space (tunnel), another ones are silly
     (PPP).
   - packet socket receives packets with pulled ll header,
     so that SOCK_RAW should push it back.

On receive:
-----------

Incoming, dev->hard_header!=NULL
113 114
   mac_header -> ll header
   data       -> data
Linus Torvalds's avatar
Linus Torvalds committed
115 116

Outgoing, dev->hard_header!=NULL
117 118
   mac_header -> ll header
   data       -> ll header
Linus Torvalds's avatar
Linus Torvalds committed
119 120

Incoming, dev->hard_header==NULL
121 122
   mac_header -> UNKNOWN position. It is very likely, that it points to ll
		 header.  PPP makes it, that is wrong, because introduce
123
		 assymetry between rx and tx paths.
124
   data       -> data
Linus Torvalds's avatar
Linus Torvalds committed
125 126

Outgoing, dev->hard_header==NULL
127 128
   mac_header -> data. ll header is still not built!
   data       -> data
Linus Torvalds's avatar
Linus Torvalds committed
129 130 131 132 133 134 135 136 137

Resume
  If dev->hard_header==NULL we are unlikely to restore sensible ll header.


On transmit:
------------

dev->hard_header != NULL
138 139
   mac_header -> ll header
   data       -> ll header
Linus Torvalds's avatar
Linus Torvalds committed
140 141

dev->hard_header == NULL (ll header is added by device, we cannot control it)
142 143
   mac_header -> data
   data       -> data
Linus Torvalds's avatar
Linus Torvalds committed
144 145 146 147 148 149 150

   We should set nh.raw on output to correct posistion,
   packet classifier depends on it.
 */

/* Private packet socket structures. */

151 152 153
/* identical to struct packet_mreq except it has
 * a longer address field.
 */
Eric Dumazet's avatar
Eric Dumazet committed
154
struct packet_mreq_max {
155 156 157 158
	int		mr_ifindex;
	unsigned short	mr_type;
	unsigned short	mr_alen;
	unsigned char	mr_address[MAX_ADDR_LEN];
Linus Torvalds's avatar
Linus Torvalds committed
159
};
160

161 162 163 164 165 166 167
union tpacket_uhdr {
	struct tpacket_hdr  *h1;
	struct tpacket2_hdr *h2;
	struct tpacket3_hdr *h3;
	void *raw;
};

168
static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u,
Johann Baudy's avatar
Johann Baudy committed
169 170
		int closing, int tx_ring);

171 172
#define V3_ALIGNMENT	(8)

173
#define BLK_HDR_LEN	(ALIGN(sizeof(struct tpacket_block_desc), V3_ALIGNMENT))
174 175 176 177 178

#define BLK_PLUS_PRIV(sz_of_priv) \
	(BLK_HDR_LEN + ALIGN((sz_of_priv), V3_ALIGNMENT))

#define PGV_FROM_VMALLOC 1
Johann Baudy's avatar
Johann Baudy committed
179

180 181 182 183 184 185 186 187
#define BLOCK_STATUS(x)	((x)->hdr.bh1.block_status)
#define BLOCK_NUM_PKTS(x)	((x)->hdr.bh1.num_pkts)
#define BLOCK_O2FP(x)		((x)->hdr.bh1.offset_to_first_pkt)
#define BLOCK_LEN(x)		((x)->hdr.bh1.blk_len)
#define BLOCK_SNUM(x)		((x)->hdr.bh1.seq_num)
#define BLOCK_O2PRIV(x)	((x)->offset_to_priv)
#define BLOCK_PRIV(x)		((void *)((char *)(x) + BLOCK_O2PRIV(x)))

Johann Baudy's avatar
Johann Baudy committed
188 189
struct packet_sock;
static int tpacket_snd(struct packet_sock *po, struct msghdr *msg);
190 191
static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
		       struct packet_type *pt, struct net_device *orig_dev);
Linus Torvalds's avatar
Linus Torvalds committed
192

193 194 195 196
static void *packet_previous_frame(struct packet_sock *po,
		struct packet_ring_buffer *rb,
		int status);
static void packet_increment_head(struct packet_ring_buffer *buff);
197 198 199
static int prb_curr_blk_in_use(struct tpacket_kbdq_core *,
			struct tpacket_block_desc *);
static void *prb_dispatch_next_block(struct tpacket_kbdq_core *,
200
			struct packet_sock *);
201
static void prb_retire_current_block(struct tpacket_kbdq_core *,
202
		struct packet_sock *, unsigned int status);
203 204 205
static int prb_queue_frozen(struct tpacket_kbdq_core *);
static void prb_open_block(struct tpacket_kbdq_core *,
		struct tpacket_block_desc *);
206
static void prb_retire_rx_blk_timer_expired(unsigned long);
207 208 209 210 211 212 213 214 215
static void _prb_refresh_rx_retire_blk_timer(struct tpacket_kbdq_core *);
static void prb_init_blk_timer(struct packet_sock *,
		struct tpacket_kbdq_core *,
		void (*func) (unsigned long));
static void prb_fill_rxhash(struct tpacket_kbdq_core *, struct tpacket3_hdr *);
static void prb_clear_rxhash(struct tpacket_kbdq_core *,
		struct tpacket3_hdr *);
static void prb_fill_vlan_info(struct tpacket_kbdq_core *,
		struct tpacket3_hdr *);
Linus Torvalds's avatar
Linus Torvalds committed
216 217
static void packet_flush_mclist(struct sock *sk);

218 219 220
struct packet_skb_cb {
	union {
		struct sockaddr_pkt pkt;
221 222 223 224 225 226 227 228
		union {
			/* Trick: alias skb original length with
			 * ll.sll_family and ll.protocol in order
			 * to save room.
			 */
			unsigned int origlen;
			struct sockaddr_ll ll;
		};
229 230 231 232
	} sa;
};

#define PACKET_SKB_CB(__skb)	((struct packet_skb_cb *)((__skb)->cb))
233

234
#define GET_PBDQC_FROM_RB(x)	((struct tpacket_kbdq_core *)(&(x)->prb_bdqc))
235
#define GET_PBLOCK_DESC(x, bid)	\
236
	((struct tpacket_block_desc *)((x)->pkbdq[(bid)].buffer))
237
#define GET_CURR_PBLOCK_DESC_FROM_CORE(x)	\
238
	((struct tpacket_block_desc *)((x)->pkbdq[(x)->kactive_blk_num].buffer))
239 240 241 242
#define GET_NEXT_PRB_BLK_NUM(x) \
	(((x)->kactive_blk_num < ((x)->knum_blocks-1)) ? \
	((x)->kactive_blk_num+1) : 0)

David S. Miller's avatar
David S. Miller committed
243 244 245
static void __fanout_unlink(struct sock *sk, struct packet_sock *po);
static void __fanout_link(struct sock *sk, struct packet_sock *po);

246 247 248 249 250
static int packet_direct_xmit(struct sk_buff *skb)
{
	struct net_device *dev = skb->dev;
	netdev_features_t features;
	struct netdev_queue *txq;
251
	int ret = NETDEV_TX_BUSY;
252 253

	if (unlikely(!netif_running(dev) ||
254 255
		     !netif_carrier_ok(dev)))
		goto drop;
256 257 258

	features = netif_skb_features(skb);
	if (skb_needs_linearize(skb, features) &&
259 260
	    __skb_linearize(skb))
		goto drop;
261

262
	txq = skb_get_tx_queue(dev, skb);
263

264 265 266
	local_bh_disable();

	HARD_TX_LOCK(dev, txq, smp_processor_id());
267
	if (!netif_xmit_frozen_or_drv_stopped(txq))
268
		ret = netdev_start_xmit(skb, dev, txq, false);
269
	HARD_TX_UNLOCK(dev, txq);
270

271 272 273
	local_bh_enable();

	if (!dev_xmit_complete(ret))
274
		kfree_skb(skb);
275

276
	return ret;
277
drop:
278
	atomic_long_inc(&dev->tx_dropped);
279 280
	kfree_skb(skb);
	return NET_XMIT_DROP;
281 282
}

283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306
static struct net_device *packet_cached_dev_get(struct packet_sock *po)
{
	struct net_device *dev;

	rcu_read_lock();
	dev = rcu_dereference(po->cached_dev);
	if (likely(dev))
		dev_hold(dev);
	rcu_read_unlock();

	return dev;
}

static void packet_cached_dev_assign(struct packet_sock *po,
				     struct net_device *dev)
{
	rcu_assign_pointer(po->cached_dev, dev);
}

static void packet_cached_dev_reset(struct packet_sock *po)
{
	RCU_INIT_POINTER(po->cached_dev, NULL);
}

307 308 309 310 311
static bool packet_use_direct_xmit(const struct packet_sock *po)
{
	return po->xmit == packet_direct_xmit;
}

312
static u16 __packet_pick_tx_queue(struct net_device *dev, struct sk_buff *skb)
313
{
314
	return (u16) raw_smp_processor_id() % dev->real_num_tx_queues;
315 316
}

317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332
static void packet_pick_tx_queue(struct net_device *dev, struct sk_buff *skb)
{
	const struct net_device_ops *ops = dev->netdev_ops;
	u16 queue_index;

	if (ops->ndo_select_queue) {
		queue_index = ops->ndo_select_queue(dev, skb, NULL,
						    __packet_pick_tx_queue);
		queue_index = netdev_cap_txqueue(dev, queue_index);
	} else {
		queue_index = __packet_pick_tx_queue(dev, skb);
	}

	skb_set_queue_mapping(skb, queue_index);
}

333 334 335 336 337 338 339
/* register_prot_hook must be invoked with the po->bind_lock held,
 * or from a context in which asynchronous accesses to the packet
 * socket is not possible (packet_create()).
 */
static void register_prot_hook(struct sock *sk)
{
	struct packet_sock *po = pkt_sk(sk);
340

341
	if (!po->running) {
342
		if (po->fanout)
David S. Miller's avatar
David S. Miller committed
343
			__fanout_link(sk, po);
344
		else
David S. Miller's avatar
David S. Miller committed
345
			dev_add_pack(&po->prot_hook);
346

347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363
		sock_hold(sk);
		po->running = 1;
	}
}

/* {,__}unregister_prot_hook() must be invoked with the po->bind_lock
 * held.   If the sync parameter is true, we will temporarily drop
 * the po->bind_lock and do a synchronize_net to make sure no
 * asynchronous packet processing paths still refer to the elements
 * of po->prot_hook.  If the sync parameter is false, it is the
 * callers responsibility to take care of this.
 */
static void __unregister_prot_hook(struct sock *sk, bool sync)
{
	struct packet_sock *po = pkt_sk(sk);

	po->running = 0;
364 365

	if (po->fanout)
David S. Miller's avatar
David S. Miller committed
366
		__fanout_unlink(sk, po);
367
	else
David S. Miller's avatar
David S. Miller committed
368
		__dev_remove_pack(&po->prot_hook);
369

370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386
	__sock_put(sk);

	if (sync) {
		spin_unlock(&po->bind_lock);
		synchronize_net();
		spin_lock(&po->bind_lock);
	}
}

static void unregister_prot_hook(struct sock *sk, bool sync)
{
	struct packet_sock *po = pkt_sk(sk);

	if (po->running)
		__unregister_prot_hook(sk, sync);
}

387
static inline struct page * __pure pgv_to_page(void *addr)
388 389 390 391 392 393
{
	if (is_vmalloc_addr(addr))
		return vmalloc_to_page(addr);
	return virt_to_page(addr);
}

Johann Baudy's avatar
Johann Baudy committed
394
static void __packet_set_status(struct packet_sock *po, void *frame, int status)
Linus Torvalds's avatar
Linus Torvalds committed
395
{
396
	union tpacket_uhdr h;
Linus Torvalds's avatar
Linus Torvalds committed
397

Johann Baudy's avatar
Johann Baudy committed
398
	h.raw = frame;
399 400
	switch (po->tp_version) {
	case TPACKET_V1:
Johann Baudy's avatar
Johann Baudy committed
401
		h.h1->tp_status = status;
402
		flush_dcache_page(pgv_to_page(&h.h1->tp_status));
403 404
		break;
	case TPACKET_V2:
Johann Baudy's avatar
Johann Baudy committed
405
		h.h2->tp_status = status;
406
		flush_dcache_page(pgv_to_page(&h.h2->tp_status));
407
		break;
408
	case TPACKET_V3:
Johann Baudy's avatar
Johann Baudy committed
409
	default:
410
		WARN(1, "TPACKET version not supported.\n");
Johann Baudy's avatar
Johann Baudy committed
411
		BUG();
412
	}
Johann Baudy's avatar
Johann Baudy committed
413 414

	smp_wmb();
415 416
}

Johann Baudy's avatar
Johann Baudy committed
417
static int __packet_get_status(struct packet_sock *po, void *frame)
418
{
419
	union tpacket_uhdr h;
420

Johann Baudy's avatar
Johann Baudy committed
421 422
	smp_rmb();

423 424 425
	h.raw = frame;
	switch (po->tp_version) {
	case TPACKET_V1:
426
		flush_dcache_page(pgv_to_page(&h.h1->tp_status));
Johann Baudy's avatar
Johann Baudy committed
427
		return h.h1->tp_status;
428
	case TPACKET_V2:
429
		flush_dcache_page(pgv_to_page(&h.h2->tp_status));
Johann Baudy's avatar
Johann Baudy committed
430
		return h.h2->tp_status;
431
	case TPACKET_V3:
Johann Baudy's avatar
Johann Baudy committed
432
	default:
433
		WARN(1, "TPACKET version not supported.\n");
Johann Baudy's avatar
Johann Baudy committed
434 435
		BUG();
		return 0;
436
	}
Linus Torvalds's avatar
Linus Torvalds committed
437
}
Johann Baudy's avatar
Johann Baudy committed
438

439 440
static __u32 tpacket_get_timestamp(struct sk_buff *skb, struct timespec *ts,
				   unsigned int flags)
441 442 443
{
	struct skb_shared_hwtstamps *shhwtstamps = skb_hwtstamps(skb);

444 445 446 447
	if (shhwtstamps &&
	    (flags & SOF_TIMESTAMPING_RAW_HARDWARE) &&
	    ktime_to_timespec_cond(shhwtstamps->hwtstamp, ts))
		return TP_STATUS_TS_RAW_HARDWARE;
448 449

	if (ktime_to_timespec_cond(skb->tstamp, ts))
450
		return TP_STATUS_TS_SOFTWARE;
451

452
	return 0;
453 454
}

455 456
static __u32 __packet_set_timestamp(struct packet_sock *po, void *frame,
				    struct sk_buff *skb)
457 458 459
{
	union tpacket_uhdr h;
	struct timespec ts;
460
	__u32 ts_status;
461

462 463
	if (!(ts_status = tpacket_get_timestamp(skb, &ts, po->tp_tstamp)))
		return 0;
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483

	h.raw = frame;
	switch (po->tp_version) {
	case TPACKET_V1:
		h.h1->tp_sec = ts.tv_sec;
		h.h1->tp_usec = ts.tv_nsec / NSEC_PER_USEC;
		break;
	case TPACKET_V2:
		h.h2->tp_sec = ts.tv_sec;
		h.h2->tp_nsec = ts.tv_nsec;
		break;
	case TPACKET_V3:
	default:
		WARN(1, "TPACKET version not supported.\n");
		BUG();
	}

	/* one flush is safe, as both fields always lie on the same cacheline */
	flush_dcache_page(pgv_to_page(&h.h1->tp_sec));
	smp_wmb();
484 485

	return ts_status;
486 487
}

Johann Baudy's avatar
Johann Baudy committed
488 489 490 491 492 493
static void *packet_lookup_frame(struct packet_sock *po,
		struct packet_ring_buffer *rb,
		unsigned int position,
		int status)
{
	unsigned int pg_vec_pos, frame_offset;
494
	union tpacket_uhdr h;
Johann Baudy's avatar
Johann Baudy committed
495 496 497 498

	pg_vec_pos = position / rb->frames_per_block;
	frame_offset = position % rb->frames_per_block;

499 500
	h.raw = rb->pg_vec[pg_vec_pos].buffer +
		(frame_offset * rb->frame_size);
Johann Baudy's avatar
Johann Baudy committed
501 502 503 504 505 506 507

	if (status != __packet_get_status(po, h.raw))
		return NULL;

	return h.raw;
}

508
static void *packet_current_frame(struct packet_sock *po,
Johann Baudy's avatar
Johann Baudy committed
509 510 511 512 513 514
		struct packet_ring_buffer *rb,
		int status)
{
	return packet_lookup_frame(po, rb, rb->head, status);
}

515
static void prb_del_retire_blk_timer(struct tpacket_kbdq_core *pkc)
516 517 518 519 520 521 522 523
{
	del_timer_sync(&pkc->retire_blk_timer);
}

static void prb_shutdown_retire_blk_timer(struct packet_sock *po,
		int tx_ring,
		struct sk_buff_head *rb_queue)
{
524
	struct tpacket_kbdq_core *pkc;
525

526 527
	pkc = tx_ring ? GET_PBDQC_FROM_RB(&po->tx_ring) :
			GET_PBDQC_FROM_RB(&po->rx_ring);
528

529
	spin_lock_bh(&rb_queue->lock);
530
	pkc->delete_blk_timer = 1;
531
	spin_unlock_bh(&rb_queue->lock);
532 533 534 535 536

	prb_del_retire_blk_timer(pkc);
}

static void prb_init_blk_timer(struct packet_sock *po,
537
		struct tpacket_kbdq_core *pkc,
538 539 540 541 542 543 544 545 546 547
		void (*func) (unsigned long))
{
	init_timer(&pkc->retire_blk_timer);
	pkc->retire_blk_timer.data = (long)po;
	pkc->retire_blk_timer.function = func;
	pkc->retire_blk_timer.expires = jiffies;
}

static void prb_setup_retire_blk_timer(struct packet_sock *po, int tx_ring)
{
548
	struct tpacket_kbdq_core *pkc;
549 550 551 552

	if (tx_ring)
		BUG();

553 554
	pkc = tx_ring ? GET_PBDQC_FROM_RB(&po->tx_ring) :
			GET_PBDQC_FROM_RB(&po->rx_ring);
555 556 557 558 559 560 561 562
	prb_init_blk_timer(po, pkc, prb_retire_rx_blk_timer_expired);
}

static int prb_calc_retire_blk_tmo(struct packet_sock *po,
				int blk_size_in_bytes)
{
	struct net_device *dev;
	unsigned int mbits = 0, msec = 0, div = 0, tmo = 0;
563 564
	struct ethtool_cmd ecmd;
	int err;
565
	u32 speed;
566

567 568 569 570
	rtnl_lock();
	dev = __dev_get_by_index(sock_net(&po->sk), po->ifindex);
	if (unlikely(!dev)) {
		rtnl_unlock();
571
		return DEFAULT_PRB_RETIRE_TOV;
572 573
	}
	err = __ethtool_get_settings(dev, &ecmd);
574
	speed = ethtool_cmd_speed(&ecmd);
575 576 577 578 579 580
	rtnl_unlock();
	if (!err) {
		/*
		 * If the link speed is so slow you don't really
		 * need to worry about perf anyways
		 */
581
		if (speed < SPEED_1000 || speed == SPEED_UNKNOWN) {
582
			return DEFAULT_PRB_RETIRE_TOV;
583 584 585
		} else {
			msec = 1;
			div = speed / 1000;
586 587 588 589 590 591 592 593 594 595 596 597 598 599 600
		}
	}

	mbits = (blk_size_in_bytes * 8) / (1024 * 1024);

	if (div)
		mbits /= div;

	tmo = mbits * msec;

	if (div)
		return tmo+1;
	return tmo;
}

601
static void prb_init_ft_ops(struct tpacket_kbdq_core *p1,
602 603 604 605 606 607 608 609 610 611
			union tpacket_req_u *req_u)
{
	p1->feature_req_word = req_u->req3.tp_feature_req_word;
}

static void init_prb_bdqc(struct packet_sock *po,
			struct packet_ring_buffer *rb,
			struct pgv *pg_vec,
			union tpacket_req_u *req_u, int tx_ring)
{
612
	struct tpacket_kbdq_core *p1 = GET_PBDQC_FROM_RB(rb);
613
	struct tpacket_block_desc *pbd;
614 615 616 617 618

	memset(p1, 0x0, sizeof(*p1));

	p1->knxt_seq_num = 1;
	p1->pkbdq = pg_vec;
619
	pbd = (struct tpacket_block_desc *)pg_vec[0].buffer;
Joe Perches's avatar
Joe Perches committed
620
	p1->pkblk_start	= pg_vec[0].buffer;
621 622 623 624 625
	p1->kblk_size = req_u->req3.tp_block_size;
	p1->knum_blocks	= req_u->req3.tp_block_nr;
	p1->hdrlen = po->tp_hdrlen;
	p1->version = po->tp_version;
	p1->last_kactive_blk_num = 0;
626
	po->stats.stats3.tp_freeze_q_cnt = 0;
627 628 629 630 631 632 633 634
	if (req_u->req3.tp_retire_blk_tov)
		p1->retire_blk_tov = req_u->req3.tp_retire_blk_tov;
	else
		p1->retire_blk_tov = prb_calc_retire_blk_tmo(po,
						req_u->req3.tp_block_size);
	p1->tov_in_jiffies = msecs_to_jiffies(p1->retire_blk_tov);
	p1->blk_sizeof_priv = req_u->req3.tp_sizeof_priv;

635
	p1->max_frame_len = p1->kblk_size - BLK_PLUS_PRIV(p1->blk_sizeof_priv);
636 637 638 639 640 641 642 643
	prb_init_ft_ops(p1, req_u);
	prb_setup_retire_blk_timer(po, tx_ring);
	prb_open_block(p1, pbd);
}

/*  Do NOT update the last_blk_num first.
 *  Assumes sk_buff_head lock is held.
 */
644
static void _prb_refresh_rx_retire_blk_timer(struct tpacket_kbdq_core *pkc)
645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676
{
	mod_timer(&pkc->retire_blk_timer,
			jiffies + pkc->tov_in_jiffies);
	pkc->last_kactive_blk_num = pkc->kactive_blk_num;
}

/*
 * Timer logic:
 * 1) We refresh the timer only when we open a block.
 *    By doing this we don't waste cycles refreshing the timer
 *	  on packet-by-packet basis.
 *
 * With a 1MB block-size, on a 1Gbps line, it will take
 * i) ~8 ms to fill a block + ii) memcpy etc.
 * In this cut we are not accounting for the memcpy time.
 *
 * So, if the user sets the 'tmo' to 10ms then the timer
 * will never fire while the block is still getting filled
 * (which is what we want). However, the user could choose
 * to close a block early and that's fine.
 *
 * But when the timer does fire, we check whether or not to refresh it.
 * Since the tmo granularity is in msecs, it is not too expensive
 * to refresh the timer, lets say every '8' msecs.
 * Either the user can set the 'tmo' or we can derive it based on
 * a) line-speed and b) block-size.
 * prb_calc_retire_blk_tmo() calculates the tmo.
 *
 */
static void prb_retire_rx_blk_timer_expired(unsigned long data)
{
	struct packet_sock *po = (struct packet_sock *)data;
677
	struct tpacket_kbdq_core *pkc = GET_PBDQC_FROM_RB(&po->rx_ring);
678
	unsigned int frozen;
679
	struct tpacket_block_desc *pbd;
680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706

	spin_lock(&po->sk.sk_receive_queue.lock);

	frozen = prb_queue_frozen(pkc);
	pbd = GET_CURR_PBLOCK_DESC_FROM_CORE(pkc);

	if (unlikely(pkc->delete_blk_timer))
		goto out;

	/* We only need to plug the race when the block is partially filled.
	 * tpacket_rcv:
	 *		lock(); increment BLOCK_NUM_PKTS; unlock()
	 *		copy_bits() is in progress ...
	 *		timer fires on other cpu:
	 *		we can't retire the current block because copy_bits
	 *		is in progress.
	 *
	 */
	if (BLOCK_NUM_PKTS(pbd)) {
		while (atomic_read(&pkc->blk_fill_in_prog)) {
			/* Waiting for skb_copy_bits to finish... */
			cpu_relax();
		}
	}

	if (pkc->last_kactive_blk_num == pkc->kactive_blk_num) {
		if (!frozen) {
707 708 709 710
			if (!BLOCK_NUM_PKTS(pbd)) {
				/* An empty block. Just refresh the timer. */
				goto refresh_timer;
			}
711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746
			prb_retire_current_block(pkc, po, TP_STATUS_BLK_TMO);
			if (!prb_dispatch_next_block(pkc, po))
				goto refresh_timer;
			else
				goto out;
		} else {
			/* Case 1. Queue was frozen because user-space was
			 *	   lagging behind.
			 */
			if (prb_curr_blk_in_use(pkc, pbd)) {
				/*
				 * Ok, user-space is still behind.
				 * So just refresh the timer.
				 */
				goto refresh_timer;
			} else {
			       /* Case 2. queue was frozen,user-space caught up,
				* now the link went idle && the timer fired.
				* We don't have a block to close.So we open this
				* block and restart the timer.
				* opening a block thaws the queue,restarts timer
				* Thawing/timer-refresh is a side effect.
				*/
				prb_open_block(pkc, pbd);
				goto out;
			}
		}
	}

refresh_timer:
	_prb_refresh_rx_retire_blk_timer(pkc);

out:
	spin_unlock(&po->sk.sk_receive_queue.lock);
}

747
static void prb_flush_block(struct tpacket_kbdq_core *pkc1,
748
		struct tpacket_block_desc *pbd1, __u32 status)
749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789
{
	/* Flush everything minus the block header */

#if ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE == 1
	u8 *start, *end;

	start = (u8 *)pbd1;

	/* Skip the block header(we know header WILL fit in 4K) */
	start += PAGE_SIZE;

	end = (u8 *)PAGE_ALIGN((unsigned long)pkc1->pkblk_end);
	for (; start < end; start += PAGE_SIZE)
		flush_dcache_page(pgv_to_page(start));

	smp_wmb();
#endif

	/* Now update the block status. */

	BLOCK_STATUS(pbd1) = status;

	/* Flush the block header */

#if ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE == 1
	start = (u8 *)pbd1;
	flush_dcache_page(pgv_to_page(start));

	smp_wmb();
#endif
}

/*
 * Side effect:
 *
 * 1) flush the block
 * 2) Increment active_blk_num
 *
 * Note:We DONT refresh the timer on purpose.
 *	Because almost always the next block will be opened.
 */
790 791
static void prb_close_block(struct tpacket_kbdq_core *pkc1,
		struct tpacket_block_desc *pbd1,
792 793 794 795 796
		struct packet_sock *po, unsigned int stat)
{
	__u32 status = TP_STATUS_USER | stat;

	struct tpacket3_hdr *last_pkt;
797
	struct tpacket_hdr_v1 *h1 = &pbd1->hdr.bh1;
798
	struct sock *sk = &po->sk;
799

800
	if (po->stats.stats3.tp_drops)
801 802 803 804 805 806 807 808 809 810
		status |= TP_STATUS_LOSING;

	last_pkt = (struct tpacket3_hdr *)pkc1->prev;
	last_pkt->tp_next_offset = 0;

	/* Get the ts of the last pkt */
	if (BLOCK_NUM_PKTS(pbd1)) {
		h1->ts_last_pkt.ts_sec = last_pkt->tp_sec;
		h1->ts_last_pkt.ts_nsec	= last_pkt->tp_nsec;
	} else {
811 812 813 814 815
		/* Ok, we tmo'd - so get the current time.
		 *
		 * It shouldn't really happen as we don't close empty
		 * blocks. See prb_retire_rx_blk_timer_expired().
		 */
816 817 818 819 820 821 822 823 824 825 826
		struct timespec ts;
		getnstimeofday(&ts);
		h1->ts_last_pkt.ts_sec = ts.tv_sec;
		h1->ts_last_pkt.ts_nsec	= ts.tv_nsec;
	}

	smp_wmb();

	/* Flush the block */
	prb_flush_block(pkc1, pbd1, status);

827 828
	sk->sk_data_ready(sk);

829 830 831
	pkc1->kactive_blk_num = GET_NEXT_PRB_BLK_NUM(pkc1);
}

832
static void prb_thaw_queue(struct tpacket_kbdq_core *pkc)
833 834 835 836 837 838 839 840 841 842 843
{
	pkc->reset_pending_on_curr_blk = 0;
}

/*
 * Side effect of opening a block:
 *
 * 1) prb_queue is thawed.
 * 2) retire_blk_timer is refreshed.
 *
 */
844 845
static void prb_open_block(struct tpacket_kbdq_core *pkc1,
	struct tpacket_block_desc *pbd1)
846 847
{
	struct timespec ts;
848
	struct tpacket_hdr_v1 *h1 = &pbd1->hdr.bh1;
849 850 851

	smp_rmb();

852 853 854
	/* We could have just memset this but we will lose the
	 * flexibility of making the priv area sticky
	 */
855

856 857 858
	BLOCK_SNUM(pbd1) = pkc1->knxt_seq_num++;
	BLOCK_NUM_PKTS(pbd1) = 0;
	BLOCK_LEN(pbd1) = BLK_PLUS_PRIV(pkc1->blk_sizeof_priv);
859

860 861 862 863
	getnstimeofday(&ts);

	h1->ts_first_pkt.ts_sec = ts.tv_sec;
	h1->ts_first_pkt.ts_nsec = ts.tv_nsec;
864

865 866 867 868 869 870 871 872 873 874 875 876 877 878
	pkc1->pkblk_start = (char *)pbd1;
	pkc1->nxt_offset = pkc1->pkblk_start + BLK_PLUS_PRIV(pkc1->blk_sizeof_priv);

	BLOCK_O2FP(pbd1) = (__u32)BLK_PLUS_PRIV(pkc1->blk_sizeof_priv);
	BLOCK_O2PRIV(pbd1) = BLK_HDR_LEN;

	pbd1->version = pkc1->version;
	pkc1->prev = pkc1->nxt_offset;
	pkc1->pkblk_end = pkc1->pkblk_start + pkc1->kblk_size;

	prb_thaw_queue(pkc1);
	_prb_refresh_rx_retire_blk_timer(pkc1);

	smp_wmb();
879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903
}

/*
 * Queue freeze logic:
 * 1) Assume tp_block_nr = 8 blocks.
 * 2) At time 't0', user opens Rx ring.
 * 3) Some time past 't0', kernel starts filling blocks starting from 0 .. 7
 * 4) user-space is either sleeping or processing block '0'.
 * 5) tpacket_rcv is currently filling block '7', since there is no space left,
 *    it will close block-7,loop around and try to fill block '0'.
 *    call-flow:
 *    __packet_lookup_frame_in_block
 *      prb_retire_current_block()
 *      prb_dispatch_next_block()
 *        |->(BLOCK_STATUS == USER) evaluates to true
 *    5.1) Since block-0 is currently in-use, we just freeze the queue.
 * 6) Now there are two cases:
 *    6.1) Link goes idle right after the queue is frozen.
 *         But remember, the last open_block() refreshed the timer.
 *         When this timer expires,it will refresh itself so that we can
 *         re-open block-0 in near future.
 *    6.2) Link is busy and keeps on receiving packets. This is a simple
 *         case and __packet_lookup_frame_in_block will check if block-0
 *         is free and can now be re-used.
 */
904
static void prb_freeze_queue(struct tpacket_kbdq_core *pkc,
905 906 907
				  struct packet_sock *po)
{
	pkc->reset_pending_on_curr_blk = 1;
908
	po->stats.stats3.tp_freeze_q_cnt++;
909 910 911 912 913 914 915 916 917 918
}

#define TOTAL_PKT_LEN_INCL_ALIGN(length) (ALIGN((length), V3_ALIGNMENT))

/*
 * If the next block is free then we will dispatch it
 * and return a good offset.
 * Else, we will freeze the queue.
 * So, caller must check the return value.
 */
919
static void *prb_dispatch_next_block(struct tpacket_kbdq_core *pkc,
920 921
		struct packet_sock *po)
{
922
	struct tpacket_block_desc *pbd;
923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943

	smp_rmb();

	/* 1. Get current block num */
	pbd = GET_CURR_PBLOCK_DESC_FROM_CORE(pkc);

	/* 2. If this block is currently in_use then freeze the queue */
	if (TP_STATUS_USER & BLOCK_STATUS(pbd)) {
		prb_freeze_queue(pkc, po);
		return NULL;
	}

	/*
	 * 3.
	 * open this block and return the offset where the first packet
	 * needs to get stored.
	 */
	prb_open_block(pkc, pbd);
	return (void *)pkc->nxt_offset;
}

944
static void prb_retire_current_block(struct tpacket_kbdq_core *pkc,
945 946
		struct packet_sock *po, unsigned int status)
{
947
	struct tpacket_block_desc *pbd = GET_CURR_PBLOCK_DESC_FROM_CORE(pkc);
948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970

	/* retire/close the current block */
	if (likely(TP_STATUS_KERNEL == BLOCK_STATUS(pbd))) {
		/*
		 * Plug the case where copy_bits() is in progress on
		 * cpu-0 and tpacket_rcv() got invoked on cpu-1, didn't
		 * have space to copy the pkt in the current block and
		 * called prb_retire_current_block()
		 *
		 * We don't need to worry about the TMO case because
		 * the timer-handler already handled this case.
		 */
		if (!(status & TP_STATUS_BLK_TMO)) {
			while (atomic_read(&pkc->blk_fill_in_prog)) {
				/* Waiting for skb_copy_bits to finish... */
				cpu_relax();
			}
		}
		prb_close_block(pkc, pbd, po, status);
		return;
	}
}

971
static int prb_curr_blk_in_use(struct tpacket_kbdq_core *pkc,
972
				      struct tpacket_block_desc *pbd)
973 974 975 976
{
	return TP_STATUS_USER & BLOCK_STATUS(pbd);
}

977
static int prb_queue_frozen(struct tpacket_kbdq_core *pkc)
978 979 980 981
{
	return pkc->reset_pending_on_curr_blk;
}

982
static void prb_clear_blk_fill_status(struct packet_ring_buffer *rb)
983
{
984
	struct tpacket_kbdq_core *pkc  = GET_PBDQC_FROM_RB(rb);
985 986 987
	atomic_dec(&pkc->blk_fill_in_prog);
}

988
static void prb_fill_rxhash(struct tpacket_kbdq_core *pkc,
989 990
			struct tpacket3_hdr *ppd)
{
991
	ppd->hv1.tp_rxhash = skb_get_hash(pkc->skb);
992 993
}

994
static void prb_clear_rxhash(struct tpacket_kbdq_core *pkc,
995 996 997 998 999
			struct tpacket3_hdr *ppd)
{
	ppd->hv1.tp_rxhash = 0;
}

1000
static void prb_fill_vlan_info(struct tpacket_kbdq_core *pkc,
1001 1002
			struct tpacket3_hdr *ppd)
{
1003 1004
	if (skb_vlan_tag_present(pkc->skb)) {
		ppd->hv1.tp_vlan_tci = skb_vlan_tag_get(pkc->skb);
1005 1006
		ppd->hv1.tp_vlan_tpid = ntohs(pkc->skb->vlan_proto);
		ppd->tp_status = TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
1007
	} else {
1008
		ppd->hv1.tp_vlan_tci = 0;
1009
		ppd->hv1.tp_vlan_tpid = 0;
1010
		ppd->tp_status = TP_STATUS_AVAILABLE;
1011 1012 1013
	}
}

1014
static void prb_run_all_ft_ops(struct tpacket_kbdq_core *pkc,
1015 1016
			struct tpacket3_hdr *ppd)
{
1017
	ppd->hv1.tp_padding = 0;
1018 1019 1020 1021 1022 1023 1024 1025
	prb_fill_vlan_info(pkc, ppd);

	if (pkc->feature_req_word & TP_FT_REQ_FILL_RXHASH)
		prb_fill_rxhash(pkc, ppd);
	else
		prb_clear_rxhash(pkc, ppd);
}

1026
static void prb_fill_curr_block(char *curr,
1027 1028
				struct tpacket_kbdq_core *pkc,
				struct tpacket_block_desc *pbd,
1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049
				unsigned int len)
{
	struct tpacket3_hdr *ppd;

	ppd  = (struct tpacket3_hdr *)curr;
	ppd->tp_next_offset = TOTAL_PKT_LEN_INCL_ALIGN(len);
	pkc->prev = curr;
	pkc->nxt_offset += TOTAL_PKT_LEN_INCL_ALIGN(len);
	BLOCK_LEN(pbd) += TOTAL_PKT_LEN_INCL_ALIGN(len);
	BLOCK_NUM_PKTS(pbd) += 1;
	atomic_inc(&pkc->blk_fill_in_prog);
	prb_run_all_ft_ops(pkc, ppd);
}

/* Assumes caller has the sk->rx_queue.lock */
static void *__packet_lookup_frame_in_block(struct packet_sock *po,
					    struct sk_buff *skb,
						int status,
					    unsigned int len
					    )
{
1050 1051
	struct tpacket_kbdq_core *pkc;
	struct tpacket_block_desc *pbd;
1052 1053
	char *curr, *end;

Joe Perches's avatar
Joe Perches committed
1054
	pkc = GET_PBDQC_FROM_RB(&po->rx_ring);
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079
	pbd = GET_CURR_PBLOCK_DESC_FROM_CORE(pkc);

	/* Queue is frozen when user space is lagging behind */
	if (prb_queue_frozen(pkc)) {
		/*
		 * Check if that last block which caused the queue to freeze,
		 * is still in_use by user-space.
		 */
		if (prb_curr_blk_in_use(pkc, pbd)) {
			/* Can't record this packet */
			return NULL;
		} else {
			/*
			 * Ok, the block was released by user-space.
			 * Now let's open that block.
			 * opening a block also thaws the queue.
			 * Thawing is a side effect.
			 */
			prb_open_block(pkc, pbd);
		}
	}

	smp_mb();
	curr = pkc->nxt_offset;
	pkc->skb = skb;
Joe Perches's avatar
Joe Perches committed
1080
	end = (char *)pbd + pkc->kblk_size;
1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105

	/* first try the current block */
	if (curr+TOTAL_PKT_LEN_INCL_ALIGN(len) < end) {
		prb_fill_curr_block(curr, pkc, pbd, len);
		return (void *)curr;
	}

	/* Ok, close the current block */
	prb_retire_current_block(pkc, po, 0);

	/* Now, try to dispatch the next block */
	curr = (char *)prb_dispatch_next_block(pkc, po);
	if (curr) {
		pbd = GET_CURR_PBLOCK_DESC_FROM_CORE(pkc);
		prb_fill_curr_block(curr, pkc, pbd, len);
		return (void *)curr;
	}

	/*
	 * No free blocks are available.user_space hasn't caught up yet.
	 * Queue was just frozen and now this packet will get dropped.
	 */
	return NULL;
}

1106
static void *packet_current_rx_frame(struct packet_sock *po,
1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121
					    struct sk_buff *skb,
					    int status, unsigned int len)
{
	char *curr = NULL;
	switch (po->tp_version) {
	case TPACKET_V1:
	case TPACKET_V2:
		curr = packet_lookup_frame(po, &po->rx_ring,
					po->rx_ring.head, status);
		return curr;
	case TPACKET_V3:
		return __packet_lookup_frame_in_block(po, skb, status, len);
	default:
		WARN(1, "TPACKET version not supported\n");
		BUG();
1122
		return NULL;
1123 1124 1125
	}
}

1126
static void *prb_lookup_block(struct packet_sock *po,
1127
				     struct packet_ring_buffer *rb,
1128
				     unsigned int idx,
1129 1130
				     int status)
{
1131
	struct tpacket_kbdq_core *pkc  = GET_PBDQC_FROM_RB(rb);
1132
	struct tpacket_block_desc *pbd = GET_PBLOCK_DESC(pkc, idx);
1133 1134 1135 1136 1137 1138

	if (status != BLOCK_STATUS(pbd))
		return NULL;
	return pbd;
}

1139
static int prb_previous_blk_num(struct packet_ring_buffer *rb)
1140 1141 1142 1143 1144 1145 1146 1147 1148 1149
{
	unsigned int prev;
	if (rb->prb_bdqc.kactive_blk_num)
		prev = rb->prb_bdqc.kactive_blk_num-1;
	else
		prev = rb->prb_bdqc.knum_blocks-1;
	return prev;
}

/* Assumes caller has held the rx_queue.lock */
1150
static void *__prb_previous_block(struct packet_sock *po,
1151 1152 1153 1154 1155 1156 1157
					 struct packet_ring_buffer *rb,
					 int status)
{
	unsigned int previous = prb_previous_blk_num(rb);
	return prb_lookup_block(po, rb, previous, status);
}

1158
static void *packet_previous_rx_frame(struct packet_sock *po,
1159 1160 1161 1162 1163 1164 1165 1166 1167
					     struct packet_ring_buffer *rb,
					     int status)
{
	if (po->tp_version <= TPACKET_V2)
		return packet_previous_frame(po, rb, status);

	return __prb_previous_block(po, rb, status);
}

1168
static void packet_increment_rx_head(struct packet_sock *po,
1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182
					    struct packet_ring_buffer *rb)
{
	switch (po->tp_version) {
	case TPACKET_V1:
	case TPACKET_V2:
		return packet_increment_head(rb);
	case TPACKET_V3:
	default:
		WARN(1, "TPACKET version not supported.\n");
		BUG();
		return;
	}
}

1183
static void *packet_previous_frame(struct packet_sock *po,
Johann Baudy's avatar
Johann Baudy committed
1184 1185 1186 1187 1188 1189 1190
		struct packet_ring_buffer *rb,
		int status)
{
	unsigned int previous = rb->head ? rb->head - 1 : rb->frame_max;
	return packet_lookup_frame(po, rb, previous, status);
}

1191
static void packet_increment_head(struct packet_ring_buffer *buff)
Johann Baudy's avatar
Johann Baudy committed
1192 1193 1194 1195
{
	buff->head = buff->head != buff->frame_max ? buff->head+1 : 0;
}

1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236
static void packet_inc_pending(struct packet_ring_buffer *rb)
{
	this_cpu_inc(*rb->pending_refcnt);
}

static void packet_dec_pending(struct packet_ring_buffer *rb)
{
	this_cpu_dec(*rb->pending_refcnt);
}

static unsigned int packet_read_pending(const struct packet_ring_buffer *rb)
{
	unsigned int refcnt = 0;
	int cpu;

	/* We don't use pending refcount in rx_ring. */
	if (rb->pending_refcnt == NULL)
		return 0;

	for_each_possible_cpu(cpu)
		refcnt += *per_cpu_ptr(rb->pending_refcnt, cpu);

	return refcnt;
}

static int packet_alloc_pending(struct packet_sock *po)
{
	po->rx_ring.pending_refcnt = NULL;

	po->tx_ring.pending_refcnt = alloc_percpu(unsigned int);
	if (unlikely(po->tx_ring.pending_refcnt == NULL))
		return -ENOBUFS;

	return 0;
}

static void packet_free_pending(struct packet_sock *po)
{
	free_percpu(po->tx_ring.pending_refcnt);
}

1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259
static bool packet_rcv_has_room(struct packet_sock *po, struct sk_buff *skb)
{
	struct sock *sk = &po->sk;
	bool has_room;

	if (po->prot_hook.func != tpacket_rcv)
		return (atomic_read(&sk->sk_rmem_alloc) + skb->truesize)
			<= sk->sk_rcvbuf;

	spin_lock(&sk->sk_receive_queue.lock);
	if (po->tp_version == TPACKET_V3)
		has_room = prb_lookup_block(po, &po->rx_ring,
					    po->rx_ring.prb_bdqc.kactive_blk_num,
					    TP_STATUS_KERNEL);
	else
		has_room = packet_lookup_frame(po, &po->rx_ring,
					       po->rx_ring.head,
					       TP_STATUS_KERNEL);
	spin_unlock(&sk->sk_receive_queue.lock);

	return has_room;
}

Linus Torvalds's avatar
Linus Torvalds committed
1260 1261
static void packet_sock_destruct(struct sock *sk)
{
1262 1263
	skb_queue_purge(&sk->sk_error_queue);

1264 1265
	WARN_ON(atomic_read(&sk->sk_rmem_alloc));
	WARN_ON(atomic_read(&sk->sk_wmem_alloc));
Linus Torvalds's avatar
Linus Torvalds committed
1266 1267

	if (!sock_flag(sk, SOCK_DEAD)) {
Eric Dumazet's avatar
Eric Dumazet committed
1268
		pr_err("Attempt to release alive packet socket: %p\n", sk);
Linus Torvalds's avatar
Linus Torvalds committed
1269 1270 1271
		return;
	}

1272
	sk_refcnt_debug_dec(sk);
Linus Torvalds's avatar
Linus Torvalds committed
1273 1274
}