Linux debugging

Check our new training course

Linux debugging, tracing, profiling & perf. analysis

Check our new training course
with Creative Commons CC-BY-SA
lecture and lab materials

Bootlin logo

Elixir Cross Referencer

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 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
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
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
182
183
184
185
186
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
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
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
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
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
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
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
677
678
679
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
707
708
709
710
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
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
/*
 * 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.
 *
 *		Implementation of the Transmission Control Protocol(TCP).
 *
 * Version:	$Id: tcp_timer.c,v 1.71 2000/01/18 08:24:19 davem Exp $
 *
 * Authors:	Ross Biro, <bir7@leland.Stanford.Edu>
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Mark Evans, <evansmp@uhura.aston.ac.uk>
 *		Corey Minyard <wf-rch!minyard@relay.EU.net>
 *		Florian La Roche, <flla@stud.uni-sb.de>
 *		Charles Hedrick, <hedrick@klinzhai.rutgers.edu>
 *		Linus Torvalds, <torvalds@cs.helsinki.fi>
 *		Alan Cox, <gw4pts@gw4pts.ampr.org>
 *		Matthew Dillon, <dillon@apollo.west.oic.com>
 *		Arnt Gulbrandsen, <agulbra@nvg.unit.no>
 *		Jorge Cwik, <jorge@laser.satlink.net>
 */

#include <net/tcp.h>

int sysctl_tcp_syn_retries = TCP_SYN_RETRIES; 
int sysctl_tcp_synack_retries = TCP_SYNACK_RETRIES; 
int sysctl_tcp_keepalive_time = TCP_KEEPALIVE_TIME;
int sysctl_tcp_keepalive_probes = TCP_KEEPALIVE_PROBES;
int sysctl_tcp_keepalive_intvl = TCP_KEEPALIVE_INTVL;
int sysctl_tcp_retries1 = TCP_RETR1;
int sysctl_tcp_retries2 = TCP_RETR2;
int sysctl_tcp_orphan_retries = TCP_ORPHAN_RETRIES;

static void tcp_retransmit_timer(unsigned long);
static void tcp_delack_timer(unsigned long);
static void tcp_probe_timer(unsigned long);
static void tcp_keepalive_timer (unsigned long data);
static void tcp_twkill(unsigned long);

const char timer_bug_msg[] = KERN_DEBUG "tcpbug: unknown timer value\n";

/*
 * Using different timers for retransmit, delayed acks and probes
 * We may wish use just one timer maintaining a list of expire jiffies 
 * to optimize.
 */

void tcp_init_xmit_timers(struct sock *sk)
{
	struct tcp_opt *tp = &sk->tp_pinfo.af_tcp;

	spin_lock_init(&sk->timer_lock);

	init_timer(&tp->retransmit_timer);
	tp->retransmit_timer.function=&tcp_retransmit_timer;
	tp->retransmit_timer.data = (unsigned long) sk;

	init_timer(&tp->delack_timer);
	tp->delack_timer.function=&tcp_delack_timer;
	tp->delack_timer.data = (unsigned long) sk;

	init_timer(&tp->probe_timer);
	tp->probe_timer.function=&tcp_probe_timer;
	tp->probe_timer.data = (unsigned long) sk;

	init_timer(&sk->timer);
	sk->timer.function=&tcp_keepalive_timer;
	sk->timer.data = (unsigned long) sk;
}

/*
 *	Reset the retransmission timer
 */
 
void tcp_reset_xmit_timer(struct sock *sk, int what, unsigned long when)
{
	struct tcp_opt *tp = &sk->tp_pinfo.af_tcp;

	spin_lock_bh(&sk->timer_lock);
	switch (what) {
	case TCP_TIME_RETRANS:
		/* When seting the transmit timer the probe timer 
		 * should not be set.
		 * The delayed ack timer can be set if we are changing the
		 * retransmit timer when removing acked frames.
		 */
		if(tp->probe_timer.prev && del_timer(&tp->probe_timer))
			__sock_put(sk);
		if (!tp->retransmit_timer.prev || !del_timer(&tp->retransmit_timer))
			sock_hold(sk);
		if (when > TCP_RTO_MAX) {
			printk(KERN_DEBUG "reset_xmit_timer sk=%p when=0x%lx, caller=%p\n", sk, when, NET_CALLER(sk));
			when = TCP_RTO_MAX;
		}
		mod_timer(&tp->retransmit_timer, jiffies+when);
		break;

	case TCP_TIME_DACK:
		if (!tp->delack_timer.prev || !del_timer(&tp->delack_timer))
			sock_hold(sk);
		mod_timer(&tp->delack_timer, jiffies+when);
		break;

	case TCP_TIME_PROBE0:
		if (!tp->probe_timer.prev || !del_timer(&tp->probe_timer))
			sock_hold(sk);
		mod_timer(&tp->probe_timer, jiffies+when);
		break;	

	default:
		printk(KERN_DEBUG "bug: unknown timer value\n");
	};
	spin_unlock_bh(&sk->timer_lock);
}

void tcp_clear_xmit_timers(struct sock *sk)
{	
	struct tcp_opt *tp = &sk->tp_pinfo.af_tcp;

	spin_lock_bh(&sk->timer_lock);
	if(tp->retransmit_timer.prev && del_timer(&tp->retransmit_timer))
		__sock_put(sk);
	if(tp->delack_timer.prev && del_timer(&tp->delack_timer))
		__sock_put(sk);
	tp->ack.blocked = 0;
	if(tp->probe_timer.prev && del_timer(&tp->probe_timer))
		__sock_put(sk);
	if(sk->timer.prev && del_timer(&sk->timer))
		__sock_put(sk);
	spin_unlock_bh(&sk->timer_lock);
}

static void tcp_write_err(struct sock *sk)
{
	sk->err = sk->err_soft ? : ETIMEDOUT;
	sk->error_report(sk);

	tcp_done(sk);
}

/* A write timeout has occurred. Process the after effects. */
static int tcp_write_timeout(struct sock *sk)
{
	struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);
	int retry_until;

	if ((1<<sk->state)&(TCPF_SYN_SENT|TCPF_SYN_RECV)) {
		if (tp->retransmits)
			dst_negative_advice(&sk->dst_cache);
		retry_until = tp->syn_retries ? : sysctl_tcp_syn_retries;
	} else {
		if (tp->retransmits >= sysctl_tcp_retries1) {
			/* NOTE. draft-ietf-tcpimpl-pmtud-01.txt requires pmtu black
			   hole detection. :-(

			   It is place to make it. It is not made. I do not want
			   to make it. It is disguisting. It does not work in any
			   case. Let me to cite the same draft, which requires for
			   us to implement this:

   "The one security concern raised by this memo is that ICMP black holes
   are often caused by over-zealous security administrators who block
   all ICMP messages.  It is vitally important that those who design and
   deploy security systems understand the impact of strict filtering on
   upper-layer protocols.  The safest web site in the world is worthless
   if most TCP implementations cannot transfer data from it.  It would
   be far nicer to have all of the black holes fixed rather than fixing
   all of the TCP implementations."

                           Golden words :-).
		   */

			dst_negative_advice(&sk->dst_cache);
		}
		retry_until = sysctl_tcp_retries2;
		if (sk->dead)
			retry_until = sysctl_tcp_orphan_retries;
	}

	if (tp->retransmits >= retry_until) {
		/* Has it gone just too far? */
		tcp_write_err(sk);
		return 1;
	}
	return 0;
}

static void tcp_delack_timer(unsigned long data)
{
	struct sock *sk = (struct sock*)data;
	struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);

	bh_lock_sock(sk);
	if (sk->lock.users) {
		/* Try again later. */
		tp->ack.blocked = 1;
		NET_INC_STATS_BH(DelayedACKLocked);
		tcp_reset_xmit_timer(sk, TCP_TIME_DACK, TCP_DELACK_MIN);
		goto out_unlock;
	}

	if (tp->ack.pending) {
		/* Delayed ACK missed: inflate ATO, leave pingpong mode */
		tp->ack.ato = min(tp->ack.ato<<1, TCP_ATO_MAX);
		tp->ack.pingpong = 0;
		tcp_send_ack(sk);
		NET_INC_STATS_BH(DelayedACKs);
	}
	TCP_CHECK_TIMER(sk);

out_unlock:
	bh_unlock_sock(sk);
	sock_put(sk);
}

static void tcp_probe_timer(unsigned long data)
{
	struct sock *sk = (struct sock*)data;
	struct tcp_opt *tp = &sk->tp_pinfo.af_tcp;
	int max_probes;

	bh_lock_sock(sk);
	if (sk->lock.users) {
		/* Try again later. */
		tcp_reset_xmit_timer(sk, TCP_TIME_PROBE0, HZ/5);
		goto out_unlock;
	}

	if (sk->state == TCP_CLOSE)
		goto out_unlock;

	if (tp->packets_out || !tp->send_head) {
		tp->probes_out = 0;
		goto out_unlock;
	}

	/* *WARNING* RFC 1122 forbids this
	 *
	 * It doesn't AFAIK, because we kill the retransmit timer -AK
	 *
	 * FIXME: We ought not to do it, Solaris 2.5 actually has fixing
	 * this behaviour in Solaris down as a bug fix. [AC]
	 *
	 * Let me to explain. probes_out is zeroed by incoming ACKs
	 * even if they advertise zero window. Hence, connection is killed only
	 * if we received no ACKs for normal connection timeout. It is not killed
	 * only because window stays zero for some time, window may be zero
	 * until armageddon and even later. We are in full accordance
	 * with RFCs, only probe timer combines both retransmission timeout
	 * and probe timeout in one bottle.				--ANK
	 */
	max_probes = sk->dead ? sysctl_tcp_orphan_retries : sysctl_tcp_retries2;

	if (tp->probes_out > max_probes) {
		tcp_write_err(sk);
	} else {
		/* Only send another probe if we didn't close things up. */
		tcp_send_probe0(sk);
		TCP_CHECK_TIMER(sk);
	}
out_unlock:
	bh_unlock_sock(sk);
	sock_put(sk);
}


/* Kill off TIME_WAIT sockets once their lifetime has expired. */
static int tcp_tw_death_row_slot = 0;
int tcp_tw_count = 0;

static struct tcp_tw_bucket *tcp_tw_death_row[TCP_TWKILL_SLOTS];
static spinlock_t tw_death_lock = SPIN_LOCK_UNLOCKED;
static struct timer_list tcp_tw_timer = { function: tcp_twkill };

static void tcp_twkill(unsigned long data)
{
	struct tcp_tw_bucket *tw;
	int killed = 0;

	/* NOTE: compare this to previous version where lock
	 * was released after detaching chain. It was racy,
	 * because tw buckets are scheduled in not serialized context
	 * in 2.3 (with netfilter), and with softnet it is common, because
	 * soft irqs are not sequenced.
	 */
	spin_lock(&tw_death_lock);

	if (tcp_tw_count == 0)
		goto out;

	while((tw = tcp_tw_death_row[tcp_tw_death_row_slot]) != NULL) {
		tcp_tw_death_row[tcp_tw_death_row_slot] = tw->next_death;
		tw->pprev_death = NULL;
		spin_unlock(&tw_death_lock);

		tcp_timewait_kill(tw);
		tcp_tw_put(tw);

		killed++;

		spin_lock(&tw_death_lock);
	}
	tcp_tw_death_row_slot =
		((tcp_tw_death_row_slot + 1) & (TCP_TWKILL_SLOTS - 1));

	if ((tcp_tw_count -= killed) != 0)
		mod_timer(&tcp_tw_timer, jiffies+TCP_TWKILL_PERIOD);
	net_statistics[smp_processor_id()*2].TimeWaited += killed;
out:
	spin_unlock(&tw_death_lock);
}

/* These are always called from BH context.  See callers in
 * tcp_input.c to verify this.
 */

/* This is for handling early-kills of TIME_WAIT sockets. */
void tcp_tw_deschedule(struct tcp_tw_bucket *tw)
{
	spin_lock(&tw_death_lock);
	if (tw->pprev_death) {
		if(tw->next_death)
			tw->next_death->pprev_death = tw->pprev_death;
		*tw->pprev_death = tw->next_death;
		tw->pprev_death = NULL;
		tcp_tw_put(tw);
		if (--tcp_tw_count == 0)
			del_timer(&tcp_tw_timer);
	}
	spin_unlock(&tw_death_lock);
}

/* Short-time timewait calendar */

static int tcp_twcal_hand = -1;
static int tcp_twcal_jiffie;
static void tcp_twcal_tick(unsigned long);
static struct timer_list tcp_twcal_timer = {NULL, NULL, 0, 0, tcp_twcal_tick,};
static struct tcp_tw_bucket *tcp_twcal_row[TCP_TW_RECYCLE_SLOTS];

void tcp_tw_schedule(struct tcp_tw_bucket *tw, int timeo)
{
	struct tcp_tw_bucket **tpp;
	int slot;

	/* timeout := RTO * 3.5
	 *
	 * 3.5 = 1+2+0.5 to wait for two retransmits.
	 *
	 * RATIONALE: if FIN arrived and we entered TIME-WAIT state,
	 * our ACK acking that FIN can be lost. If N subsequent retransmitted
	 * FINs (or previous seqments) are lost (probability of such event
	 * is p^(N+1), where p is probability to lose single packet and
	 * time to detect the loss is about RTO*(2^N - 1) with exponential
	 * backoff). Normal timewait length is calculated so, that we
	 * waited at least for one retransmitted FIN (maximal RTO is 120sec).
	 * [ BTW Linux. following BSD, violates this requirement waiting
	 *   only for 60sec, we should wait at least for 240 secs.
	 *   Well, 240 consumes too much of resources 8)
	 * ]
	 * This interval is not reduced to catch old duplicate and
	 * responces to our wandering segments living for two MSLs.
	 * However, if we use PAWS to detect
	 * old duplicates, we can reduce the interval to bounds required
	 * by RTO, rather than MSL. So, if peer understands PAWS, we
	 * kill tw bucket after 3.5*RTO (it is important that this number
	 * is greater than TS tick!) and detect old duplicates with help
	 * of PAWS.
	 */
	slot = (timeo + (1<<TCP_TW_RECYCLE_TICK) - 1) >> TCP_TW_RECYCLE_TICK;

	spin_lock(&tw_death_lock);

	/* Unlink it, if it was scheduled */
	if (tw->pprev_death) {
		if(tw->next_death)
			tw->next_death->pprev_death = tw->pprev_death;
		*tw->pprev_death = tw->next_death;
		tw->pprev_death = NULL;
		tcp_tw_count--;
	} else
		atomic_inc(&tw->refcnt);

	if (slot >= TCP_TW_RECYCLE_SLOTS) {
		/* Schedule to slow timer */
		if (timeo >= TCP_TIMEWAIT_LEN) {
			slot = TCP_TWKILL_SLOTS-1;
		} else {
			slot = (timeo + TCP_TWKILL_PERIOD-1) / TCP_TWKILL_PERIOD;
			if (slot >= TCP_TWKILL_SLOTS)
				slot = TCP_TWKILL_SLOTS-1;
		}
		tw->ttd = jiffies + timeo;
		slot = (tcp_tw_death_row_slot + slot) & (TCP_TWKILL_SLOTS - 1);
		tpp = &tcp_tw_death_row[slot];
	} else {
		tw->ttd = jiffies + (slot<<TCP_TW_RECYCLE_TICK);

		if (tcp_twcal_hand < 0) {
			tcp_twcal_hand = 0;
			tcp_twcal_jiffie = jiffies;
			tcp_twcal_timer.expires = tcp_twcal_jiffie + (slot<<TCP_TW_RECYCLE_TICK);
			add_timer(&tcp_twcal_timer);
		} else {
			if ((long)(tcp_twcal_timer.expires - jiffies) > (slot<<TCP_TW_RECYCLE_TICK))
				mod_timer(&tcp_twcal_timer, jiffies + (slot<<TCP_TW_RECYCLE_TICK));
			slot = (tcp_twcal_hand + slot)&(TCP_TW_RECYCLE_SLOTS-1);
		}
		tpp = &tcp_twcal_row[slot];
	}

	if((tw->next_death = *tpp) != NULL)
		(*tpp)->pprev_death = &tw->next_death;
	*tpp = tw;
	tw->pprev_death = tpp;

	if (tcp_tw_count++ == 0)
		mod_timer(&tcp_tw_timer, jiffies+TCP_TWKILL_PERIOD);
	spin_unlock(&tw_death_lock);
}

void tcp_twcal_tick(unsigned long dummy)
{
	int n, slot;
	unsigned long j;
	unsigned long now = jiffies;
	int killed = 0;
	int adv = 0;

	spin_lock(&tw_death_lock);
	if (tcp_twcal_hand < 0)
		goto out;

	slot = tcp_twcal_hand;
	j = tcp_twcal_jiffie;

	for (n=0; n<TCP_TW_RECYCLE_SLOTS; n++) {
		if ((long)(j - now) <= 0) {
			struct tcp_tw_bucket *tw;

			while((tw = tcp_twcal_row[slot]) != NULL) {
				tcp_twcal_row[slot] = tw->next_death;
				tw->pprev_death = NULL;

				tcp_timewait_kill(tw);
				tcp_tw_put(tw);
				killed++;
			}
		} else {
			if (!adv) {
				adv = 1;
				tcp_twcal_jiffie = j;
				tcp_twcal_hand = slot;
			}

			if (tcp_twcal_row[slot] != NULL) {
				mod_timer(&tcp_twcal_timer, j);
				goto out;
			}
		}
		j += (1<<TCP_TW_RECYCLE_TICK);
		slot = (slot+1)&(TCP_TW_RECYCLE_SLOTS-1);
	}
	tcp_twcal_hand = -1;

out:
	if ((tcp_tw_count -= killed) == 0)
		del_timer(&tcp_tw_timer);
	net_statistics[smp_processor_id()*2].TimeWaitKilled += killed;
	spin_unlock(&tw_death_lock);
}


/*
 *	The TCP retransmit timer.
 */

static void tcp_retransmit_timer(unsigned long data)
{
	struct sock *sk = (struct sock*)data;
	struct tcp_opt *tp = &sk->tp_pinfo.af_tcp;

	bh_lock_sock(sk);
	if (sk->lock.users) {
		/* Try again later */  
		tcp_reset_xmit_timer(sk, TCP_TIME_RETRANS, HZ/20);
		goto out_unlock;
	}

	if (sk->state == TCP_CLOSE || tp->packets_out == 0)
		goto out_unlock;

	BUG_TRAP(!skb_queue_empty(&sk->write_queue));

	if (tcp_write_timeout(sk))
		goto out_unlock;

	/* RFC 2018, clear all 'sacked' flags in retransmission queue,
	 * the sender may have dropped out of order frames and we must
	 * send them out should this timer fire on us.
	 */
	if(tp->sack_ok) {
		struct sk_buff *skb = skb_peek(&sk->write_queue);

		while((skb != NULL) &&
		      (skb != tp->send_head) &&
		      (skb != (struct sk_buff *)&sk->write_queue)) {
			TCP_SKB_CB(skb)->sacked &=
				~(TCPCB_SACKED_ACKED | TCPCB_SACKED_RETRANS);
			skb = skb->next;
		}
	}

	/* Retransmission. */
	tp->retrans_head = NULL;
	tp->rexmt_done = 0;
	tp->fackets_out = 0;
	tp->retrans_out = 0;
	if (tp->retransmits == 0) {
		/* Remember window where we lost:
		 * "one half of the current window but at least 2 segments"
		 *
		 * Here "current window" means the effective one, which
		 * means it must be an accurate representation of our current
		 * sending rate _and_ the snd_wnd.
		 */
		tp->snd_ssthresh = tcp_recalc_ssthresh(tp);
		tp->snd_cwnd_cnt = 0;
		tp->snd_cwnd = 1;
	}

	tp->dup_acks = 0;
	tp->high_seq = tp->snd_nxt;
	if (tcp_retransmit_skb(sk, skb_peek(&sk->write_queue)) > 0) {
		/* Retransmission failed because of local congestion,
		 * do not backoff.
		 */
		if (!tp->retransmits)
			tp->retransmits=1;
		tcp_reset_xmit_timer(sk, TCP_TIME_RETRANS,
				     min(tp->rto, TCP_RESOURCE_PROBE_INTERVAL));
		TCP_CHECK_TIMER(sk);
		goto out_unlock;
	}

	/* Increase the timeout each time we retransmit.  Note that
	 * we do not increase the rtt estimate.  rto is initialized
	 * from rtt, but increases here.  Jacobson (SIGCOMM 88) suggests
	 * that doubling rto each time is the least we can get away with.
	 * In KA9Q, Karn uses this for the first few times, and then
	 * goes to quadratic.  netBSD doubles, but only goes up to *64,
	 * and clamps at 1 to 64 sec afterwards.  Note that 120 sec is
	 * defined in the protocol as the maximum possible RTT.  I guess
	 * we'll have to use something other than TCP to talk to the
	 * University of Mars.
	 *
	 * PAWS allows us longer timeouts and large windows, so once
	 * implemented ftp to mars will work nicely. We will have to fix
	 * the 120 second clamps though!
	 */
	tp->backoff++;
	tp->retransmits++;
	tp->rto = min(tp->rto << 1, TCP_RTO_MAX);
	tcp_reset_xmit_timer(sk, TCP_TIME_RETRANS, tp->rto);
	TCP_CHECK_TIMER(sk);

out_unlock:
	bh_unlock_sock(sk);
	sock_put(sk);
}

/*
 *	Timer for listening sockets
 */

static void tcp_synack_timer(struct sock *sk)
{
	struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);
	struct tcp_listen_opt *lopt = tp->listen_opt;
	int max_retries = tp->syn_retries ? : sysctl_tcp_synack_retries;
	int thresh = max_retries;
	unsigned long now = jiffies;
	struct open_request **reqp, *req;
	int i, budget;

	if (lopt == NULL || lopt->qlen == 0)
		return;

	/* Normally all the openreqs are young and become mature
	 * (i.e. converted to established socket) for first timeout.
	 * If synack was not acknowledged for 3 seconds, it means
	 * one of the following things: synack was lost, ack was lost,
	 * rtt is high or nobody planned to ack (i.e. synflood).
	 * When server is a bit loaded, queue is populated with old
	 * open requests, reducing effective size of queue.
	 * When server is well loaded, queue size reduces to zero
	 * after several minutes of work. It is not synflood,
	 * it is normal operation. The solution is pruning
	 * too old entries overriding normal timeout, when
	 * situation becomes dangerous.
	 *
	 * Essentially, we reserve half of room for young
	 * embrions; and abort old ones without pity, if old
	 * ones are about to clog our table.
	 */
	if (lopt->qlen>>(lopt->max_qlen_log-1)) {
		int young = (lopt->qlen_young<<1);

		while (thresh > 2) {
			if (lopt->qlen < young)
				break;
			thresh--;
			young <<= 1;
		}
	}

	if (tp->defer_accept)
		max_retries = tp->defer_accept;

	budget = 2*(TCP_SYNQ_HSIZE/(TCP_TIMEOUT_INIT/TCP_SYNQ_INTERVAL));
	i = lopt->clock_hand;

	do {
		reqp=&lopt->syn_table[i];
		while ((req = *reqp) != NULL) {
			if ((long)(now - req->expires) >= 0) {
				if ((req->retrans < thresh ||
				     (req->acked && req->retrans < max_retries))
				    && !req->class->rtx_syn_ack(sk, req, NULL)) {
					unsigned long timeo;

					if (req->retrans++ == 0)
						lopt->qlen_young--;
					timeo = min((TCP_TIMEOUT_INIT << req->retrans),
						    TCP_RTO_MAX);
					req->expires = now + timeo;
					reqp = &req->dl_next;
					continue;
				}

				/* Drop this request */
				write_lock(&tp->syn_wait_lock);
				*reqp = req->dl_next;
				write_unlock(&tp->syn_wait_lock);
				lopt->qlen--;
				if (req->retrans == 0)
					lopt->qlen_young--;
				tcp_openreq_free(req);
			}
			reqp = &req->dl_next;
		}

		i = (i+1)&(TCP_SYNQ_HSIZE-1);

	} while (--budget > 0);

	lopt->clock_hand = i;

	if (lopt->qlen)
		tcp_reset_keepalive_timer(sk, TCP_SYNQ_INTERVAL);
}

void tcp_delete_keepalive_timer (struct sock *sk)
{
	spin_lock_bh(&sk->timer_lock);
	if (sk->timer.prev && del_timer (&sk->timer))
		__sock_put(sk);
	spin_unlock_bh(&sk->timer_lock);
}

void tcp_reset_keepalive_timer (struct sock *sk, unsigned long len)
{
	spin_lock_bh(&sk->timer_lock);
	if(!sk->timer.prev || !del_timer(&sk->timer))
		sock_hold(sk);
	mod_timer(&sk->timer, jiffies+len);
	spin_unlock_bh(&sk->timer_lock);
}

void tcp_set_keepalive(struct sock *sk, int val)
{
	if ((1<<sk->state)&(TCPF_CLOSE|TCPF_LISTEN))
		return;

	if (val && !sk->keepopen)
		tcp_reset_keepalive_timer(sk, keepalive_time_when(&sk->tp_pinfo.af_tcp));
	else if (!val)
		tcp_delete_keepalive_timer(sk);
}


static void tcp_keepalive_timer (unsigned long data)
{
	struct sock *sk = (struct sock *) data;
	struct tcp_opt *tp = &sk->tp_pinfo.af_tcp;
	__u32 elapsed;

	/* Only process if socket is not in use. */
	bh_lock_sock(sk);
	if (sk->lock.users) {
		/* Try again later. */ 
		tcp_reset_keepalive_timer (sk, HZ/20);
		goto out;
	}

	if (sk->state == TCP_LISTEN) {
		tcp_synack_timer(sk);
		goto out;
	}

	if (sk->state == TCP_FIN_WAIT2 && sk->dead) {
		if (tp->linger2 >= 0) {
			int tmo = tcp_fin_time(tp) - TCP_TIMEWAIT_LEN;

			if (tmo > 0) {
				tcp_time_wait(sk, TCP_FIN_WAIT2, tmo);
				goto out;
			}
		}
		tcp_send_active_reset(sk, GFP_ATOMIC);
		goto death;
	}

	if (!sk->keepopen || sk->state == TCP_CLOSE)
		goto out;

	elapsed = keepalive_time_when(tp);

	/* It is alive without keepalive 8) */
	if (tp->packets_out || tp->send_head)
		goto resched;

	elapsed = tcp_time_stamp - tp->rcv_tstamp;

	if (elapsed >= keepalive_time_when(tp)) {
		if ((!tp->keepalive_probes && tp->probes_out >= sysctl_tcp_keepalive_probes) ||
		     (tp->keepalive_probes && tp->probes_out >= tp->keepalive_probes)) {
			tcp_send_active_reset(sk, GFP_ATOMIC);
			tcp_write_err(sk);
			goto out;
		}
		if (tcp_write_wakeup(sk) <= 0) {
			tp->probes_out++;
			elapsed = keepalive_intvl_when(tp);
		} else {
			/* If keepalive was lost due to local congestion,
			 * try harder.
			 */
			elapsed = TCP_RESOURCE_PROBE_INTERVAL;
		}
	} else {
		/* It is tp->rcv_tstamp + keepalive_time_when(tp) */
		elapsed = keepalive_time_when(tp) - elapsed;
	}

	TCP_CHECK_TIMER(sk);

resched:
	tcp_reset_keepalive_timer (sk, elapsed);
	goto out;

death:	
	tcp_done(sk);

out:
	bh_unlock_sock(sk);
	sock_put(sk);
}