git.meshlink.io Git - utcp/blob - utcp.c

   1 /*
   2     utcp.c -- Userspace TCP
   3     Copyright (C) 2014-2017 Guus Sliepen <guus@tinc-vpn.org>
   4
   5     This program is free software; you can redistribute it and/or modify
   6     it under the terms of the GNU General Public License as published by
   7     the Free Software Foundation; either version 2 of the License, or
   8     (at your option) any later version.
   9
  10     This program is distributed in the hope that it will be useful,
  11     but WITHOUT ANY WARRANTY; without even the implied warranty of
  12     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  13     GNU General Public License for more details.
  14
  15     You should have received a copy of the GNU General Public License along
  16     with this program; if not, write to the Free Software Foundation, Inc.,
  17     51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
  18 */
  19
  20 #define _GNU_SOURCE
  21
  22 #include <assert.h>
  23 #include <errno.h>
  24 #include <stdio.h>
  25 #include <stdlib.h>
  26 #include <stdint.h>
  27 #include <stdbool.h>
  28 #include <string.h>
  29 #include <unistd.h>
  30 #include <time.h>
  31
  32 #include "utcp_priv.h"
  33
  34 #ifndef EBADMSG
  35 #define EBADMSG         104
  36 #endif
  37
  38 #ifndef SHUT_RDWR
  39 #define SHUT_RDWR 2
  40 #endif
  41
  42 #ifdef poll
  43 #undef poll
  44 #endif
  45
  46 static void timespec_sub(const struct timespec *a, const struct timespec *b, struct timespec *r) {
  47         r->tv_sec = a->tv_sec - b->tv_sec;
  48         r->tv_nsec = a->tv_nsec - b->tv_nsec;
  49
  50         if(r->tv_nsec < 0) {
  51                 r->tv_sec--, r->tv_nsec += NSEC_PER_SEC;
  52         }
  53 }
  54
  55 static int32_t timespec_diff_usec(const struct timespec *a, const struct timespec *b) {
  56         int64_t diff = (a->tv_sec - b->tv_sec) * 1000000000 + a->tv_sec - b->tv_sec;
  57         return diff / 1000;
  58 }
  59
  60 static bool timespec_lt(const struct timespec *a, const struct timespec *b) {
  61         if(a->tv_sec == b->tv_sec) {
  62                 return a->tv_nsec < b->tv_nsec;
  63         } else {
  64                 return a->tv_sec < b->tv_sec;
  65         }
  66 }
  67
  68 static void timespec_clear(struct timespec *a) {
  69         a->tv_sec = 0;
  70 }
  71
  72 static bool timespec_isset(const struct timespec *a) {
  73         return a->tv_sec;
  74 }
  75
  76 static long CLOCK_GRANULARITY; // usec
  77
  78 static inline size_t min(size_t a, size_t b) {
  79         return a < b ? a : b;
  80 }
  81
  82 static inline size_t max(size_t a, size_t b) {
  83         return a > b ? a : b;
  84 }
  85
  86 #ifdef UTCP_DEBUG
  87 #include <stdarg.h>
  88
  89 #ifndef UTCP_DEBUG_DATALEN
  90 #define UTCP_DEBUG_DATALEN 20
  91 #endif
  92
  93 #ifndef UTCP_CLOCK
  94 #if defined(CLOCK_MONOTONIC_RAW) && defined(__x86_64__)
  95 #define UTCP_CLOCK CLOCK_MONOTONIC_RAW
  96 #else
  97 #define UTCP_CLOCK CLOCK_MONOTONIC
  98 #endif
  99 #endif
 100
 101 static void debug(struct utcp_connection *c, const char *format, ...) {
 102         struct timespec tv;
 103         char buf[1024];
 104         int len;
 105
 106         clock_gettime(CLOCK_REALTIME, &tv);
 107         len = snprintf(buf, sizeof(buf), "%ld.%06lu %u:%u ", (long)tv.tv_sec, tv.tv_nsec / 1000, c ? c->src : 0, c ? c->dst : 0);
 108         va_list ap;
 109         va_start(ap, format);
 110         len += vsnprintf(buf + len, sizeof(buf) - len, format, ap);
 111         va_end(ap);
 112
 113         if(len > 0 && (size_t)len < sizeof(buf)) {
 114                 fwrite(buf, len, 1, stderr);
 115         }
 116 }
 117
 118 static void print_packet(struct utcp_connection *c, const char *dir, const void *pkt, size_t len) {
 119         struct hdr hdr;
 120
 121         if(len < sizeof(hdr)) {
 122                 debug(c, "%s: short packet (%lu bytes)\n", dir, (unsigned long)len);
 123                 return;
 124         }
 125
 126         memcpy(&hdr, pkt, sizeof(hdr));
 127
 128         uint32_t datalen;
 129
 130         if(len > sizeof(hdr)) {
 131                 datalen = min(len - sizeof(hdr), UTCP_DEBUG_DATALEN);
 132         } else {
 133                 datalen = 0;
 134         }
 135
 136
 137         const uint8_t *data = (uint8_t *)pkt + sizeof(hdr);
 138         char str[datalen * 2 + 1];
 139         char *p = str;
 140
 141         for(uint32_t i = 0; i < datalen; i++) {
 142                 *p++ = "0123456789ABCDEF"[data[i] >> 4];
 143                 *p++ = "0123456789ABCDEF"[data[i] & 15];
 144         }
 145
 146         *p = 0;
 147
 148         debug(c, "%s: len %lu src %u dst %u seq %u ack %u wnd %u aux %x ctl %s%s%s%s data %s\n",
 149               dir, (unsigned long)len, hdr.src, hdr.dst, hdr.seq, hdr.ack, hdr.wnd, hdr.aux,
 150               hdr.ctl & SYN ? "SYN" : "",
 151               hdr.ctl & RST ? "RST" : "",
 152               hdr.ctl & FIN ? "FIN" : "",
 153               hdr.ctl & ACK ? "ACK" : "",
 154               str
 155              );
 156 }
 157
 158 static void debug_cwnd(struct utcp_connection *c) {
 159         debug(c, "snd.cwnd %u snd.ssthresh %u\n", c->snd.cwnd, ~c->snd.ssthresh ? c->snd.ssthresh : 0);
 160 }
 161 #else
 162 #define debug(...) do {} while(0)
 163 #define print_packet(...) do {} while(0)
 164 #define debug_cwnd(...) do {} while(0)
 165 #endif
 166
 167 static void set_state(struct utcp_connection *c, enum state state) {
 168         c->state = state;
 169
 170         if(state == ESTABLISHED) {
 171                 timespec_clear(&c->conn_timeout);
 172         }
 173
 174         debug(c, "state %s\n", strstate[state]);
 175 }
 176
 177 static bool fin_wanted(struct utcp_connection *c, uint32_t seq) {
 178         if(seq != c->snd.last) {
 179                 return false;
 180         }
 181
 182         switch(c->state) {
 183         case FIN_WAIT_1:
 184         case CLOSING:
 185         case LAST_ACK:
 186                 return true;
 187
 188         default:
 189                 return false;
 190         }
 191 }
 192
 193 static bool is_reliable(struct utcp_connection *c) {
 194         return c->flags & UTCP_RELIABLE;
 195 }
 196
 197 static int32_t seqdiff(uint32_t a, uint32_t b) {
 198         return a - b;
 199 }
 200
 201 // Buffer functions
 202 static bool buffer_wraps(struct buffer *buf) {
 203         return buf->size - buf->offset < buf->used;
 204 }
 205
 206 static bool buffer_resize(struct buffer *buf, uint32_t newsize) {
 207         char *newdata = realloc(buf->data, newsize);
 208
 209         if(!newdata) {
 210                 return false;
 211         }
 212
 213         buf->data = newdata;
 214
 215         if(buffer_wraps(buf)) {
 216                 // Shift the right part of the buffer until it hits the end of the new buffer.
 217                 // Old situation:
 218                 // [345......012]
 219                 // New situation:
 220                 // [345.........|........012]
 221                 uint32_t tailsize = buf->size - buf->offset;
 222                 uint32_t newoffset = newsize - tailsize;
 223                 memmove(buf + newoffset, buf + buf->offset, tailsize);
 224                 buf->offset = newoffset;
 225         }
 226
 227         buf->size = newsize;
 228         return true;
 229 }
 230
 231 // Store data into the buffer
 232 static ssize_t buffer_put_at(struct buffer *buf, size_t offset, const void *data, size_t len) {
 233         debug(NULL, "buffer_put_at %lu %lu %lu\n", (unsigned long)buf->used, (unsigned long)offset, (unsigned long)len);
 234
 235         // Ensure we don't store more than maxsize bytes in total
 236         size_t required = offset + len;
 237
 238         if(required > buf->maxsize) {
 239                 if(offset >= buf->maxsize) {
 240                         return 0;
 241                 }
 242
 243                 len = buf->maxsize - offset;
 244                 required = buf->maxsize;
 245         }
 246
 247         // Check if we need to resize the buffer
 248         if(required > buf->size) {
 249                 size_t newsize = buf->size;
 250
 251                 if(!newsize) {
 252                         newsize = 4096;
 253                 }
 254
 255                 do {
 256                         newsize *= 2;
 257                 } while(newsize < required);
 258
 259                 if(newsize > buf->maxsize) {
 260                         newsize = buf->maxsize;
 261                 }
 262
 263                 if(!buffer_resize(buf, newsize)) {
 264                         return -1;
 265                 }
 266         }
 267
 268         uint32_t realoffset = buf->offset + offset;
 269
 270         if(buf->size - buf->offset < offset) {
 271                 // The offset wrapped
 272                 realoffset -= buf->size;
 273         }
 274
 275         if(buf->size - realoffset < len) {
 276                 // The new chunk of data must be wrapped
 277                 memcpy(buf->data + realoffset, data, buf->size - realoffset);
 278                 memcpy(buf->data, (char *)data + buf->size - realoffset, len - (buf->size - realoffset));
 279         } else {
 280                 memcpy(buf->data + realoffset, data, len);
 281         }
 282
 283         if(required > buf->used) {
 284                 buf->used = required;
 285         }
 286
 287         return len;
 288 }
 289
 290 static ssize_t buffer_put(struct buffer *buf, const void *data, size_t len) {
 291         return buffer_put_at(buf, buf->used, data, len);
 292 }
 293
 294 // Copy data from the buffer without removing it.
 295 static ssize_t buffer_copy(struct buffer *buf, void *data, size_t offset, size_t len) {
 296         // Ensure we don't copy more than is actually stored in the buffer
 297         if(offset >= buf->used) {
 298                 return 0;
 299         }
 300
 301         if(buf->used - offset < len) {
 302                 len = buf->used - offset;
 303         }
 304
 305         uint32_t realoffset = buf->offset + offset;
 306
 307         if(buf->size - buf->offset < offset) {
 308                 // The offset wrapped
 309                 realoffset -= buf->size;
 310         }
 311
 312         if(buf->size - realoffset < len) {
 313                 // The data is wrapped
 314                 memcpy(data, buf->data + realoffset, buf->size - realoffset);
 315                 memcpy((char *)data + buf->size - realoffset, buf->data, len - (buf->size - realoffset));
 316         } else {
 317                 memcpy(data, buf->data + realoffset, len);
 318         }
 319
 320         return len;
 321 }
 322
 323 // Discard data from the buffer.
 324 static ssize_t buffer_discard(struct buffer *buf, size_t len) {
 325         if(buf->used < len) {
 326                 len = buf->used;
 327         }
 328
 329         if(buf->size - buf->offset < len) {
 330                 buf->offset -= buf->size;
 331         }
 332
 333         buf->offset += len;
 334         buf->used -= len;
 335
 336         return len;
 337 }
 338
 339 static bool buffer_set_size(struct buffer *buf, uint32_t minsize, uint32_t maxsize) {
 340         if(maxsize < minsize) {
 341                 maxsize = minsize;
 342         }
 343
 344         buf->maxsize = maxsize;
 345
 346         return buf->size >= minsize || buffer_resize(buf, minsize);
 347 }
 348
 349 static void buffer_exit(struct buffer *buf) {
 350         free(buf->data);
 351         memset(buf, 0, sizeof(*buf));
 352 }
 353
 354 static uint32_t buffer_free(const struct buffer *buf) {
 355         return buf->maxsize - buf->used;
 356 }
 357
 358 // Connections are stored in a sorted list.
 359 // This gives O(log(N)) lookup time, O(N log(N)) insertion time and O(N) deletion time.
 360
 361 static int compare(const void *va, const void *vb) {
 362         assert(va && vb);
 363
 364         const struct utcp_connection *a = *(struct utcp_connection **)va;
 365         const struct utcp_connection *b = *(struct utcp_connection **)vb;
 366
 367         assert(a && b);
 368         assert(a->src && b->src);
 369
 370         int c = (int)a->src - (int)b->src;
 371
 372         if(c) {
 373                 return c;
 374         }
 375
 376         c = (int)a->dst - (int)b->dst;
 377         return c;
 378 }
 379
 380 static struct utcp_connection *find_connection(const struct utcp *utcp, uint16_t src, uint16_t dst) {
 381         if(!utcp->nconnections) {
 382                 return NULL;
 383         }
 384
 385         struct utcp_connection key = {
 386                 .src = src,
 387                 .dst = dst,
 388         }, *keyp = &key;
 389         struct utcp_connection **match = bsearch(&keyp, utcp->connections, utcp->nconnections, sizeof(*utcp->connections), compare);
 390         return match ? *match : NULL;
 391 }
 392
 393 static void free_connection(struct utcp_connection *c) {
 394         struct utcp *utcp = c->utcp;
 395         struct utcp_connection **cp = bsearch(&c, utcp->connections, utcp->nconnections, sizeof(*utcp->connections), compare);
 396
 397         assert(cp);
 398
 399         int i = cp - utcp->connections;
 400         memmove(cp, cp + 1, (utcp->nconnections - i - 1) * sizeof(*cp));
 401         utcp->nconnections--;
 402
 403         buffer_exit(&c->rcvbuf);
 404         buffer_exit(&c->sndbuf);
 405         free(c);
 406 }
 407
 408 static struct utcp_connection *allocate_connection(struct utcp *utcp, uint16_t src, uint16_t dst) {
 409         // Check whether this combination of src and dst is free
 410
 411         if(src) {
 412                 if(find_connection(utcp, src, dst)) {
 413                         errno = EADDRINUSE;
 414                         return NULL;
 415                 }
 416         } else { // If src == 0, generate a random port number with the high bit set
 417                 if(utcp->nconnections >= 32767) {
 418                         errno = ENOMEM;
 419                         return NULL;
 420                 }
 421
 422                 src = rand() | 0x8000;
 423
 424                 while(find_connection(utcp, src, dst)) {
 425                         src++;
 426                 }
 427         }
 428
 429         // Allocate memory for the new connection
 430
 431         if(utcp->nconnections >= utcp->nallocated) {
 432                 if(!utcp->nallocated) {
 433                         utcp->nallocated = 4;
 434                 } else {
 435                         utcp->nallocated *= 2;
 436                 }
 437
 438                 struct utcp_connection **new_array = realloc(utcp->connections, utcp->nallocated * sizeof(*utcp->connections));
 439
 440                 if(!new_array) {
 441                         return NULL;
 442                 }
 443
 444                 utcp->connections = new_array;
 445         }
 446
 447         struct utcp_connection *c = calloc(1, sizeof(*c));
 448
 449         if(!c) {
 450                 return NULL;
 451         }
 452
 453         if(!buffer_set_size(&c->sndbuf, DEFAULT_SNDBUFSIZE, DEFAULT_MAXSNDBUFSIZE)) {
 454                 free(c);
 455                 return NULL;
 456         }
 457
 458         if(!buffer_set_size(&c->rcvbuf, DEFAULT_RCVBUFSIZE, DEFAULT_MAXRCVBUFSIZE)) {
 459                 buffer_exit(&c->sndbuf);
 460                 free(c);
 461                 return NULL;
 462         }
 463
 464         // Fill in the details
 465
 466         c->src = src;
 467         c->dst = dst;
 468 #ifdef UTCP_DEBUG
 469         c->snd.iss = 0;
 470 #else
 471         c->snd.iss = rand();
 472 #endif
 473         c->snd.una = c->snd.iss;
 474         c->snd.nxt = c->snd.iss + 1;
 475         c->snd.last = c->snd.nxt;
 476         c->snd.cwnd = (utcp->mss > 2190 ? 2 : utcp->mss > 1095 ? 3 : 4) * utcp->mss;
 477         c->snd.ssthresh = ~0;
 478         debug_cwnd(c);
 479         c->utcp = utcp;
 480
 481         // Add it to the sorted list of connections
 482
 483         utcp->connections[utcp->nconnections++] = c;
 484         qsort(utcp->connections, utcp->nconnections, sizeof(*utcp->connections), compare);
 485
 486         return c;
 487 }
 488
 489 static inline uint32_t absdiff(uint32_t a, uint32_t b) {
 490         if(a > b) {
 491                 return a - b;
 492         } else {
 493                 return b - a;
 494         }
 495 }
 496
 497 // Update RTT variables. See RFC 6298.
 498 static void update_rtt(struct utcp_connection *c, uint32_t rtt) {
 499         if(!rtt) {
 500                 debug(c, "invalid rtt\n");
 501                 return;
 502         }
 503
 504         struct utcp *utcp = c->utcp;
 505
 506         if(!utcp->srtt) {
 507                 utcp->srtt = rtt;
 508                 utcp->rttvar = rtt / 2;
 509         } else {
 510                 utcp->rttvar = (utcp->rttvar * 3 + absdiff(utcp->srtt, rtt)) / 4;
 511                 utcp->srtt = (utcp->srtt * 7 + rtt) / 8;
 512         }
 513
 514         utcp->rto = utcp->srtt + max(4 * utcp->rttvar, CLOCK_GRANULARITY);
 515
 516         if(utcp->rto > MAX_RTO) {
 517                 utcp->rto = MAX_RTO;
 518         }
 519
 520         debug(c, "rtt %u srtt %u rttvar %u rto %u\n", rtt, utcp->srtt, utcp->rttvar, utcp->rto);
 521 }
 522
 523 static void start_retransmit_timer(struct utcp_connection *c) {
 524         clock_gettime(UTCP_CLOCK, &c->rtrx_timeout);
 525
 526         uint32_t rto = c->utcp->rto;
 527
 528         while(rto > USEC_PER_SEC) {
 529                 c->rtrx_timeout.tv_sec++;
 530                 rto -= USEC_PER_SEC;
 531         }
 532
 533         c->rtrx_timeout.tv_nsec += c->utcp->rto * 1000;
 534
 535         if(c->rtrx_timeout.tv_nsec >= NSEC_PER_SEC) {
 536                 c->rtrx_timeout.tv_nsec -= NSEC_PER_SEC;
 537                 c->rtrx_timeout.tv_sec++;
 538         }
 539
 540         debug(c, "rtrx_timeout %ld.%06lu\n", c->rtrx_timeout.tv_sec, c->rtrx_timeout.tv_nsec);
 541 }
 542
 543 static void stop_retransmit_timer(struct utcp_connection *c) {
 544         timespec_clear(&c->rtrx_timeout);
 545         debug(c, "rtrx_timeout cleared\n");
 546 }
 547
 548 struct utcp_connection *utcp_connect_ex(struct utcp *utcp, uint16_t dst, utcp_recv_t recv, void *priv, uint32_t flags) {
 549         struct utcp_connection *c = allocate_connection(utcp, 0, dst);
 550
 551         if(!c) {
 552                 return NULL;
 553         }
 554
 555         assert((flags & ~0x1f) == 0);
 556
 557         c->flags = flags;
 558         c->recv = recv;
 559         c->priv = priv;
 560
 561         struct {
 562                 struct hdr hdr;
 563                 uint8_t init[4];
 564         } pkt;
 565
 566         pkt.hdr.src = c->src;
 567         pkt.hdr.dst = c->dst;
 568         pkt.hdr.seq = c->snd.iss;
 569         pkt.hdr.ack = 0;
 570         pkt.hdr.wnd = c->rcvbuf.maxsize;
 571         pkt.hdr.ctl = SYN;
 572         pkt.hdr.aux = 0x0101;
 573         pkt.init[0] = 1;
 574         pkt.init[1] = 0;
 575         pkt.init[2] = 0;
 576         pkt.init[3] = flags & 0x7;
 577
 578         set_state(c, SYN_SENT);
 579
 580         print_packet(c, "send", &pkt, sizeof(pkt));
 581         utcp->send(utcp, &pkt, sizeof(pkt));
 582
 583         clock_gettime(UTCP_CLOCK, &c->conn_timeout);
 584         c->conn_timeout.tv_sec += utcp->timeout;
 585
 586         start_retransmit_timer(c);
 587
 588         return c;
 589 }
 590
 591 struct utcp_connection *utcp_connect(struct utcp *utcp, uint16_t dst, utcp_recv_t recv, void *priv) {
 592         return utcp_connect_ex(utcp, dst, recv, priv, UTCP_TCP);
 593 }
 594
 595 void utcp_accept(struct utcp_connection *c, utcp_recv_t recv, void *priv) {
 596         if(c->reapable || c->state != SYN_RECEIVED) {
 597                 debug(c, "accept() called on invalid connection in state %s\n", c, strstate[c->state]);
 598                 return;
 599         }
 600
 601         debug(c, "accepted %p %p\n", c, recv, priv);
 602         c->recv = recv;
 603         c->priv = priv;
 604         set_state(c, ESTABLISHED);
 605 }
 606
 607 static void ack(struct utcp_connection *c, bool sendatleastone) {
 608         int32_t left = seqdiff(c->snd.last, c->snd.nxt);
 609         int32_t cwndleft = min(c->snd.cwnd, c->snd.wnd) - seqdiff(c->snd.nxt, c->snd.una);
 610
 611         assert(left >= 0);
 612
 613         if(cwndleft <= 0) {
 614                 left = 0;
 615         } else if(cwndleft < left) {
 616                 left = cwndleft;
 617
 618                 if(!sendatleastone || cwndleft > c->utcp->mss) {
 619                         left -= left % c->utcp->mss;
 620                 }
 621         }
 622
 623         debug(c, "cwndleft %d left %d\n", cwndleft, left);
 624
 625         if(!left && !sendatleastone) {
 626                 return;
 627         }
 628
 629         struct {
 630                 struct hdr hdr;
 631                 uint8_t data[];
 632         } *pkt = c->utcp->pkt;
 633
 634         pkt->hdr.src = c->src;
 635         pkt->hdr.dst = c->dst;
 636         pkt->hdr.ack = c->rcv.nxt;
 637         pkt->hdr.wnd = c->rcvbuf.maxsize;
 638         pkt->hdr.ctl = ACK;
 639         pkt->hdr.aux = 0;
 640
 641         do {
 642                 uint32_t seglen = left > c->utcp->mss ? c->utcp->mss : left;
 643                 pkt->hdr.seq = c->snd.nxt;
 644
 645                 buffer_copy(&c->sndbuf, pkt->data, seqdiff(c->snd.nxt, c->snd.una), seglen);
 646
 647                 c->snd.nxt += seglen;
 648                 left -= seglen;
 649
 650                 if(seglen && fin_wanted(c, c->snd.nxt)) {
 651                         seglen--;
 652                         pkt->hdr.ctl |= FIN;
 653                 }
 654
 655                 if(!c->rtt_start.tv_sec) {
 656                         // Start RTT measurement
 657                         clock_gettime(UTCP_CLOCK, &c->rtt_start);
 658                         c->rtt_seq = pkt->hdr.seq + seglen;
 659                         debug(c, "starting RTT measurement, expecting ack %u\n", c->rtt_seq);
 660                 }
 661
 662                 print_packet(c, "send", pkt, sizeof(pkt->hdr) + seglen);
 663                 c->utcp->send(c->utcp, pkt, sizeof(pkt->hdr) + seglen);
 664         } while(left);
 665 }
 666
 667 ssize_t utcp_send(struct utcp_connection *c, const void *data, size_t len) {
 668         if(c->reapable) {
 669                 debug(c, "send() called on closed connection\n");
 670                 errno = EBADF;
 671                 return -1;
 672         }
 673
 674         switch(c->state) {
 675         case CLOSED:
 676         case LISTEN:
 677                 debug(c, "send() called on unconnected connection\n");
 678                 errno = ENOTCONN;
 679                 return -1;
 680
 681         case SYN_SENT:
 682         case SYN_RECEIVED:
 683         case ESTABLISHED:
 684         case CLOSE_WAIT:
 685                 break;
 686
 687         case FIN_WAIT_1:
 688         case FIN_WAIT_2:
 689         case CLOSING:
 690         case LAST_ACK:
 691         case TIME_WAIT:
 692                 debug(c, "send() called on closed connection\n");
 693                 errno = EPIPE;
 694                 return -1;
 695         }
 696
 697         // Exit early if we have nothing to send.
 698
 699         if(!len) {
 700                 return 0;
 701         }
 702
 703         if(!data) {
 704                 errno = EFAULT;
 705                 return -1;
 706         }
 707
 708         // Check if we need to be able to buffer all data
 709
 710         if(c->flags & UTCP_NO_PARTIAL) {
 711                 if(len > buffer_free(&c->sndbuf)) {
 712                         if(len > c->sndbuf.maxsize) {
 713                                 errno = EMSGSIZE;
 714                                 return -1;
 715                         } else {
 716                                 errno = EWOULDBLOCK;
 717                                 return 0;
 718                         }
 719                 }
 720         }
 721
 722         // Add data to send buffer.
 723
 724         if(is_reliable(c) || (c->state != SYN_SENT && c->state != SYN_RECEIVED)) {
 725                 len = buffer_put(&c->sndbuf, data, len);
 726         } else {
 727                 return 0;
 728         }
 729
 730         if(len <= 0) {
 731                 if(is_reliable(c)) {
 732                         errno = EWOULDBLOCK;
 733                         return 0;
 734                 } else {
 735                         return len;
 736                 }
 737         }
 738
 739         c->snd.last += len;
 740
 741         // Don't send anything yet if the connection has not fully established yet
 742
 743         if(c->state == SYN_SENT || c->state == SYN_RECEIVED) {
 744                 return len;
 745         }
 746
 747         ack(c, false);
 748
 749         if(!is_reliable(c)) {
 750                 c->snd.una = c->snd.nxt = c->snd.last;
 751                 buffer_discard(&c->sndbuf, c->sndbuf.used);
 752         }
 753
 754         if(is_reliable(c) && !timespec_isset(&c->rtrx_timeout)) {
 755                 start_retransmit_timer(c);
 756         }
 757
 758         if(is_reliable(c) && !timespec_isset(&c->conn_timeout)) {
 759                 clock_gettime(UTCP_CLOCK, &c->conn_timeout);
 760                 c->conn_timeout.tv_sec += c->utcp->timeout;
 761         }
 762
 763         return len;
 764 }
 765
 766 static void swap_ports(struct hdr *hdr) {
 767         uint16_t tmp = hdr->src;
 768         hdr->src = hdr->dst;
 769         hdr->dst = tmp;
 770 }
 771
 772 static void fast_retransmit(struct utcp_connection *c) {
 773         if(c->state == CLOSED || c->snd.last == c->snd.una) {
 774                 debug(c, "fast_retransmit() called but nothing to retransmit!\n");
 775                 return;
 776         }
 777
 778         struct utcp *utcp = c->utcp;
 779
 780         struct {
 781                 struct hdr hdr;
 782                 uint8_t data[];
 783         } *pkt;
 784
 785         pkt = malloc(c->utcp->mtu);
 786
 787         if(!pkt) {
 788                 return;
 789         }
 790
 791         pkt->hdr.src = c->src;
 792         pkt->hdr.dst = c->dst;
 793         pkt->hdr.wnd = c->rcvbuf.maxsize;
 794         pkt->hdr.aux = 0;
 795
 796         switch(c->state) {
 797         case ESTABLISHED:
 798         case FIN_WAIT_1:
 799         case CLOSE_WAIT:
 800         case CLOSING:
 801         case LAST_ACK:
 802                 // Send unacked data again.
 803                 pkt->hdr.seq = c->snd.una;
 804                 pkt->hdr.ack = c->rcv.nxt;
 805                 pkt->hdr.ctl = ACK;
 806                 uint32_t len = min(seqdiff(c->snd.last, c->snd.una), utcp->mss);
 807
 808                 if(fin_wanted(c, c->snd.una + len)) {
 809                         len--;
 810                         pkt->hdr.ctl |= FIN;
 811                 }
 812
 813                 buffer_copy(&c->sndbuf, pkt->data, 0, len);
 814                 print_packet(c, "rtrx", pkt, sizeof(pkt->hdr) + len);
 815                 utcp->send(utcp, pkt, sizeof(pkt->hdr) + len);
 816                 break;
 817
 818         default:
 819                 break;
 820         }
 821
 822         free(pkt);
 823 }
 824
 825 static void retransmit(struct utcp_connection *c) {
 826         if(c->state == CLOSED || c->snd.last == c->snd.una) {
 827                 debug(c, "retransmit() called but nothing to retransmit!\n");
 828                 stop_retransmit_timer(c);
 829                 return;
 830         }
 831
 832         struct utcp *utcp = c->utcp;
 833
 834         struct {
 835                 struct hdr hdr;
 836                 uint8_t data[];
 837         } *pkt = c->utcp->pkt;
 838
 839         pkt->hdr.src = c->src;
 840         pkt->hdr.dst = c->dst;
 841         pkt->hdr.wnd = c->rcvbuf.maxsize;
 842         pkt->hdr.aux = 0;
 843
 844         switch(c->state) {
 845         case SYN_SENT:
 846                 // Send our SYN again
 847                 pkt->hdr.seq = c->snd.iss;
 848                 pkt->hdr.ack = 0;
 849                 pkt->hdr.ctl = SYN;
 850                 pkt->hdr.aux = 0x0101;
 851                 pkt->data[0] = 1;
 852                 pkt->data[1] = 0;
 853                 pkt->data[2] = 0;
 854                 pkt->data[3] = c->flags & 0x7;
 855                 print_packet(c, "rtrx", pkt, sizeof(pkt->hdr) + 4);
 856                 utcp->send(utcp, pkt, sizeof(pkt->hdr) + 4);
 857                 break;
 858
 859         case SYN_RECEIVED:
 860                 // Send SYNACK again
 861                 pkt->hdr.seq = c->snd.nxt;
 862                 pkt->hdr.ack = c->rcv.nxt;
 863                 pkt->hdr.ctl = SYN | ACK;
 864                 print_packet(c, "rtrx", pkt, sizeof(pkt->hdr));
 865                 utcp->send(utcp, pkt, sizeof(pkt->hdr));
 866                 break;
 867
 868         case ESTABLISHED:
 869         case FIN_WAIT_1:
 870         case CLOSE_WAIT:
 871         case CLOSING:
 872         case LAST_ACK:
 873                 // Send unacked data again.
 874                 pkt->hdr.seq = c->snd.una;
 875                 pkt->hdr.ack = c->rcv.nxt;
 876                 pkt->hdr.ctl = ACK;
 877                 uint32_t len = min(seqdiff(c->snd.last, c->snd.una), utcp->mss);
 878
 879                 if(fin_wanted(c, c->snd.una + len)) {
 880                         len--;
 881                         pkt->hdr.ctl |= FIN;
 882                 }
 883
 884                 // RFC 5681 slow start after timeout
 885                 uint32_t flightsize = seqdiff(c->snd.nxt, c->snd.una);
 886                 c->snd.ssthresh = max(flightsize / 2, utcp->mss * 2); // eq. 4
 887                 c->snd.cwnd = utcp->mss;
 888                 debug_cwnd(c);
 889
 890                 buffer_copy(&c->sndbuf, pkt->data, 0, len);
 891                 print_packet(c, "rtrx", pkt, sizeof(pkt->hdr) + len);
 892                 utcp->send(utcp, pkt, sizeof(pkt->hdr) + len);
 893
 894                 c->snd.nxt = c->snd.una + len;
 895                 break;
 896
 897         case CLOSED:
 898         case LISTEN:
 899         case TIME_WAIT:
 900         case FIN_WAIT_2:
 901                 // We shouldn't need to retransmit anything in this state.
 902 #ifdef UTCP_DEBUG
 903                 abort();
 904 #endif
 905                 stop_retransmit_timer(c);
 906                 goto cleanup;
 907         }
 908
 909         start_retransmit_timer(c);
 910         utcp->rto *= 2;
 911
 912         if(utcp->rto > MAX_RTO) {
 913                 utcp->rto = MAX_RTO;
 914         }
 915
 916         c->rtt_start.tv_sec = 0; // invalidate RTT timer
 917         c->dupack = 0; // cancel any ongoing fast recovery
 918
 919 cleanup:
 920         return;
 921 }
 922
 923 /* Update receive buffer and SACK entries after consuming data.
 924  *
 925  * Situation:
 926  *
 927  * |.....0000..1111111111.....22222......3333|
 928  * |---------------^
 929  *
 930  * 0..3 represent the SACK entries. The ^ indicates up to which point we want
 931  * to remove data from the receive buffer. The idea is to substract "len"
 932  * from the offset of all the SACK entries, and then remove/cut down entries
 933  * that are shifted to before the start of the receive buffer.
 934  *
 935  * There are three cases:
 936  * - the SACK entry is after ^, in that case just change the offset.
 937  * - the SACK entry starts before and ends after ^, so we have to
 938  *   change both its offset and size.
 939  * - the SACK entry is completely before ^, in that case delete it.
 940  */
 941 static void sack_consume(struct utcp_connection *c, size_t len) {
 942         debug(c, "sack_consume %lu\n", (unsigned long)len);
 943
 944         if(len > c->rcvbuf.used) {
 945                 debug(c, "all SACK entries consumed\n");
 946                 c->sacks[0].len = 0;
 947                 return;
 948         }
 949
 950         buffer_discard(&c->rcvbuf, len);
 951
 952         for(int i = 0; i < NSACKS && c->sacks[i].len;) {
 953                 if(len < c->sacks[i].offset) {
 954                         c->sacks[i].offset -= len;
 955                         i++;
 956                 } else if(len < c->sacks[i].offset + c->sacks[i].len) {
 957                         c->sacks[i].len -= len - c->sacks[i].offset;
 958                         c->sacks[i].offset = 0;
 959                         i++;
 960                 } else {
 961                         if(i < NSACKS - 1) {
 962                                 memmove(&c->sacks[i], &c->sacks[i + 1], (NSACKS - 1 - i) * sizeof(c->sacks)[i]);
 963                                 c->sacks[NSACKS - 1].len = 0;
 964                         } else {
 965                                 c->sacks[i].len = 0;
 966                                 break;
 967                         }
 968                 }
 969         }
 970
 971         for(int i = 0; i < NSACKS && c->sacks[i].len; i++) {
 972                 debug(c, "SACK[%d] offset %u len %u\n", i, c->sacks[i].offset, c->sacks[i].len);
 973         }
 974 }
 975
 976 static void handle_out_of_order(struct utcp_connection *c, uint32_t offset, const void *data, size_t len) {
 977         debug(c, "out of order packet, offset %u\n", offset);
 978         // Packet loss or reordering occured. Store the data in the buffer.
 979         ssize_t rxd = buffer_put_at(&c->rcvbuf, offset, data, len);
 980
 981         if(rxd < 0 || (size_t)rxd < len) {
 982                 abort();
 983         }
 984
 985         // Make note of where we put it.
 986         for(int i = 0; i < NSACKS; i++) {
 987                 if(!c->sacks[i].len) { // nothing to merge, add new entry
 988                         debug(c, "new SACK entry %d\n", i);
 989                         c->sacks[i].offset = offset;
 990                         c->sacks[i].len = rxd;
 991                         break;
 992                 } else if(offset < c->sacks[i].offset) {
 993                         if(offset + rxd < c->sacks[i].offset) { // insert before
 994                                 if(!c->sacks[NSACKS - 1].len) { // only if room left
 995                                         debug(c, "insert SACK entry at %d\n", i);
 996                                         memmove(&c->sacks[i + 1], &c->sacks[i], (NSACKS - i - 1) * sizeof(c->sacks)[i]);
 997                                         c->sacks[i].offset = offset;
 998                                         c->sacks[i].len = rxd;
 999                                 } else {
1000                                         debug(c, "SACK entries full, dropping packet\n");
1001                                 }
1002
1003                                 break;
1004                         } else { // merge
1005                                 debug(c, "merge with start of SACK entry at %d\n", i);
1006                                 c->sacks[i].offset = offset;
1007                                 break;
1008                         }
1009                 } else if(offset <= c->sacks[i].offset + c->sacks[i].len) {
1010                         if(offset + rxd > c->sacks[i].offset + c->sacks[i].len) { // merge
1011                                 debug(c, "merge with end of SACK entry at %d\n", i);
1012                                 c->sacks[i].len = offset + rxd - c->sacks[i].offset;
1013                                 // TODO: handle potential merge with next entry
1014                         }
1015
1016                         break;
1017                 }
1018         }
1019
1020         for(int i = 0; i < NSACKS && c->sacks[i].len; i++) {
1021                 debug(c, "SACK[%d] offset %u len %u\n", i, c->sacks[i].offset, c->sacks[i].len);
1022         }
1023 }
1024
1025 static void handle_in_order(struct utcp_connection *c, const void *data, size_t len) {
1026         // Check if we can process out-of-order data now.
1027         if(c->sacks[0].len && len >= c->sacks[0].offset) { // TODO: handle overlap with second SACK
1028                 debug(c, "incoming packet len %lu connected with SACK at %u\n", (unsigned long)len, c->sacks[0].offset);
1029                 buffer_put_at(&c->rcvbuf, 0, data, len); // TODO: handle return value
1030                 len = max(len, c->sacks[0].offset + c->sacks[0].len);
1031                 data = c->rcvbuf.data;
1032         }
1033
1034         if(c->recv) {
1035                 ssize_t rxd = c->recv(c, data, len);
1036
1037                 if(rxd < 0 || (size_t)rxd != len) {
1038                         // TODO: handle the application not accepting all data.
1039                         abort();
1040                 }
1041         }
1042
1043         if(c->rcvbuf.used) {
1044                 sack_consume(c, len);
1045         }
1046
1047         c->rcv.nxt += len;
1048 }
1049
1050
1051 static void handle_incoming_data(struct utcp_connection *c, uint32_t seq, const void *data, size_t len) {
1052         if(!is_reliable(c)) {
1053                 c->recv(c, data, len);
1054                 c->rcv.nxt = seq + len;
1055                 return;
1056         }
1057
1058         uint32_t offset = seqdiff(seq, c->rcv.nxt);
1059
1060         if(offset + len > c->rcvbuf.maxsize) {
1061                 abort();
1062         }
1063
1064         if(offset) {
1065                 handle_out_of_order(c, offset, data, len);
1066         } else {
1067                 handle_in_order(c, data, len);
1068         }
1069 }
1070
1071
1072 ssize_t utcp_recv(struct utcp *utcp, const void *data, size_t len) {
1073         const uint8_t *ptr = data;
1074
1075         if(!utcp) {
1076                 errno = EFAULT;
1077                 return -1;
1078         }
1079
1080         if(!len) {
1081                 return 0;
1082         }
1083
1084         if(!data) {
1085                 errno = EFAULT;
1086                 return -1;
1087         }
1088
1089         // Drop packets smaller than the header
1090
1091         struct hdr hdr;
1092
1093         if(len < sizeof(hdr)) {
1094                 print_packet(NULL, "recv", data, len);
1095                 errno = EBADMSG;
1096                 return -1;
1097         }
1098
1099         // Make a copy from the potentially unaligned data to a struct hdr
1100
1101         memcpy(&hdr, ptr, sizeof(hdr));
1102
1103         // Try to match the packet to an existing connection
1104
1105         struct utcp_connection *c = find_connection(utcp, hdr.dst, hdr.src);
1106         print_packet(c, "recv", data, len);
1107
1108         // Process the header
1109
1110         ptr += sizeof(hdr);
1111         len -= sizeof(hdr);
1112
1113         // Drop packets with an unknown CTL flag
1114
1115         if(hdr.ctl & ~(SYN | ACK | RST | FIN)) {
1116                 print_packet(NULL, "recv", data, len);
1117                 errno = EBADMSG;
1118                 return -1;
1119         }
1120
1121         // Check for auxiliary headers
1122
1123         const uint8_t *init = NULL;
1124
1125         uint16_t aux = hdr.aux;
1126
1127         while(aux) {
1128                 size_t auxlen = 4 * (aux >> 8) & 0xf;
1129                 uint8_t auxtype = aux & 0xff;
1130
1131                 if(len < auxlen) {
1132                         errno = EBADMSG;
1133                         return -1;
1134                 }
1135
1136                 switch(auxtype) {
1137                 case AUX_INIT:
1138                         if(!(hdr.ctl & SYN) || auxlen != 4) {
1139                                 errno = EBADMSG;
1140                                 return -1;
1141                         }
1142
1143                         init = ptr;
1144                         break;
1145
1146                 default:
1147                         errno = EBADMSG;
1148                         return -1;
1149                 }
1150
1151                 len -= auxlen;
1152                 ptr += auxlen;
1153
1154                 if(!(aux & 0x800)) {
1155                         break;
1156                 }
1157
1158                 if(len < 2) {
1159                         errno = EBADMSG;
1160                         return -1;
1161                 }
1162
1163                 memcpy(&aux, ptr, 2);
1164                 len -= 2;
1165                 ptr += 2;
1166         }
1167
1168         bool has_data = len || (hdr.ctl & (SYN | FIN));
1169
1170         // Is it for a new connection?
1171
1172         if(!c) {
1173                 // Ignore RST packets
1174
1175                 if(hdr.ctl & RST) {
1176                         return 0;
1177                 }
1178
1179                 // Is it a SYN packet and are we LISTENing?
1180
1181                 if(hdr.ctl & SYN && !(hdr.ctl & ACK) && utcp->accept) {
1182                         // If we don't want to accept it, send a RST back
1183                         if((utcp->pre_accept && !utcp->pre_accept(utcp, hdr.dst))) {
1184                                 len = 1;
1185                                 goto reset;
1186                         }
1187
1188                         // Try to allocate memory, otherwise send a RST back
1189                         c = allocate_connection(utcp, hdr.dst, hdr.src);
1190
1191                         if(!c) {
1192                                 len = 1;
1193                                 goto reset;
1194                         }
1195
1196                         // Parse auxilliary information
1197                         if(init) {
1198                                 if(init[0] < 1) {
1199                                         len = 1;
1200                                         goto reset;
1201                                 }
1202
1203                                 c->flags = init[3] & 0x7;
1204                         } else {
1205                                 c->flags = UTCP_TCP;
1206                         }
1207
1208 synack:
1209                         // Return SYN+ACK, go to SYN_RECEIVED state
1210                         c->snd.wnd = hdr.wnd;
1211                         c->rcv.irs = hdr.seq;
1212                         c->rcv.nxt = c->rcv.irs + 1;
1213                         set_state(c, SYN_RECEIVED);
1214
1215                         struct {
1216                                 struct hdr hdr;
1217                                 uint8_t data[4];
1218                         } pkt;
1219
1220                         pkt.hdr.src = c->src;
1221                         pkt.hdr.dst = c->dst;
1222                         pkt.hdr.ack = c->rcv.irs + 1;
1223                         pkt.hdr.seq = c->snd.iss;
1224                         pkt.hdr.wnd = c->rcvbuf.maxsize;
1225                         pkt.hdr.ctl = SYN | ACK;
1226
1227                         if(init) {
1228                                 pkt.hdr.aux = 0x0101;
1229                                 pkt.data[0] = 1;
1230                                 pkt.data[1] = 0;
1231                                 pkt.data[2] = 0;
1232                                 pkt.data[3] = c->flags & 0x7;
1233                                 print_packet(c, "send", &pkt, sizeof(hdr) + 4);
1234                                 utcp->send(utcp, &pkt, sizeof(hdr) + 4);
1235                         } else {
1236                                 pkt.hdr.aux = 0;
1237                                 print_packet(c, "send", &pkt, sizeof(hdr));
1238                                 utcp->send(utcp, &pkt, sizeof(hdr));
1239                         }
1240                 } else {
1241                         // No, we don't want your packets, send a RST back
1242                         len = 1;
1243                         goto reset;
1244                 }
1245
1246                 return 0;
1247         }
1248
1249         debug(c, "state %s\n", strstate[c->state]);
1250
1251         // In case this is for a CLOSED connection, ignore the packet.
1252         // TODO: make it so incoming packets can never match a CLOSED connection.
1253
1254         if(c->state == CLOSED) {
1255                 debug(c, "got packet for closed connection\n");
1256                 return 0;
1257         }
1258
1259         // It is for an existing connection.
1260
1261         // 1. Drop invalid packets.
1262
1263         // 1a. Drop packets that should not happen in our current state.
1264
1265         switch(c->state) {
1266         case SYN_SENT:
1267         case SYN_RECEIVED:
1268         case ESTABLISHED:
1269         case FIN_WAIT_1:
1270         case FIN_WAIT_2:
1271         case CLOSE_WAIT:
1272         case CLOSING:
1273         case LAST_ACK:
1274         case TIME_WAIT:
1275                 break;
1276
1277         default:
1278 #ifdef UTCP_DEBUG
1279                 abort();
1280 #endif
1281                 break;
1282         }
1283
1284         // 1b. Discard data that is not in our receive window.
1285
1286         if(is_reliable(c)) {
1287                 bool acceptable;
1288
1289                 if(c->state == SYN_SENT) {
1290                         acceptable = true;
1291                 } else if(len == 0) {
1292                         acceptable = seqdiff(hdr.seq, c->rcv.nxt) >= 0;
1293                 } else {
1294                         int32_t rcv_offset = seqdiff(hdr.seq, c->rcv.nxt);
1295
1296                         // cut already accepted front overlapping
1297                         if(rcv_offset < 0) {
1298                                 acceptable = len > (size_t) - rcv_offset;
1299
1300                                 if(acceptable) {
1301                                         ptr -= rcv_offset;
1302                                         len += rcv_offset;
1303                                         hdr.seq -= rcv_offset;
1304                                 }
1305                         } else {
1306                                 acceptable = seqdiff(hdr.seq, c->rcv.nxt) >= 0 && seqdiff(hdr.seq, c->rcv.nxt) + len <= c->rcvbuf.maxsize;
1307                         }
1308                 }
1309
1310                 if(!acceptable) {
1311                         debug(c, "packet not acceptable, %u <= %u + %lu < %u\n", c->rcv.nxt, hdr.seq, (unsigned long)len, c->rcv.nxt + c->rcvbuf.maxsize);
1312
1313                         // Ignore unacceptable RST packets.
1314                         if(hdr.ctl & RST) {
1315                                 return 0;
1316                         }
1317
1318                         // Otherwise, continue processing.
1319                         len = 0;
1320                 }
1321         } else {
1322 #if UTCP_DEBUG
1323                 int32_t rcv_offset = seqdiff(hdr.seq, c->rcv.nxt);
1324
1325                 if(rcv_offset) {
1326                         debug(c, "packet out of order, offset %u bytes", rcv_offset);
1327                 }
1328
1329                 if(rcv_offset >= 0) {
1330                         c->rcv.nxt = hdr.seq + len;
1331                 }
1332
1333 #endif
1334         }
1335
1336         c->snd.wnd = hdr.wnd; // TODO: move below
1337
1338         // 1c. Drop packets with an invalid ACK.
1339         // ackno should not roll back, and it should also not be bigger than what we ever could have sent
1340         // (= snd.una + c->sndbuf.used).
1341
1342         if(!is_reliable(c)) {
1343                 if(hdr.ack != c->snd.last && c->state >= ESTABLISHED) {
1344                         hdr.ack = c->snd.una;
1345                 }
1346         }
1347
1348         if(hdr.ctl & ACK && (seqdiff(hdr.ack, c->snd.last) > 0 || seqdiff(hdr.ack, c->snd.una) < 0)) {
1349                 debug(c, "packet ack seqno out of range, %u <= %u < %u\n", c->snd.una, hdr.ack, c->snd.una + c->sndbuf.used);
1350
1351                 // Ignore unacceptable RST packets.
1352                 if(hdr.ctl & RST) {
1353                         return 0;
1354                 }
1355
1356                 goto reset;
1357         }
1358
1359         // 2. Handle RST packets
1360
1361         if(hdr.ctl & RST) {
1362                 switch(c->state) {
1363                 case SYN_SENT:
1364                         if(!(hdr.ctl & ACK)) {
1365                                 return 0;
1366                         }
1367
1368                         // The peer has refused our connection.
1369                         set_state(c, CLOSED);
1370                         errno = ECONNREFUSED;
1371
1372                         if(c->recv) {
1373                                 c->recv(c, NULL, 0);
1374                         }
1375
1376                         if(c->poll && !c->reapable) {
1377                                 c->poll(c, 0);
1378                         }
1379
1380                         return 0;
1381
1382                 case SYN_RECEIVED:
1383                         if(hdr.ctl & ACK) {
1384                                 return 0;
1385                         }
1386
1387                         // We haven't told the application about this connection yet. Silently delete.
1388                         free_connection(c);
1389                         return 0;
1390
1391                 case ESTABLISHED:
1392                 case FIN_WAIT_1:
1393                 case FIN_WAIT_2:
1394                 case CLOSE_WAIT:
1395                         if(hdr.ctl & ACK) {
1396                                 return 0;
1397                         }
1398
1399                         // The peer has aborted our connection.
1400                         set_state(c, CLOSED);
1401                         errno = ECONNRESET;
1402
1403                         if(c->recv) {
1404                                 c->recv(c, NULL, 0);
1405                         }
1406
1407                         if(c->poll && !c->reapable) {
1408                                 c->poll(c, 0);
1409                         }
1410
1411                         return 0;
1412
1413                 case CLOSING:
1414                 case LAST_ACK:
1415                 case TIME_WAIT:
1416                         if(hdr.ctl & ACK) {
1417                                 return 0;
1418                         }
1419
1420                         // As far as the application is concerned, the connection has already been closed.
1421                         // If it has called utcp_close() already, we can immediately free this connection.
1422                         if(c->reapable) {
1423                                 free_connection(c);
1424                                 return 0;
1425                         }
1426
1427                         // Otherwise, immediately move to the CLOSED state.
1428                         set_state(c, CLOSED);
1429                         return 0;
1430
1431                 default:
1432 #ifdef UTCP_DEBUG
1433                         abort();
1434 #endif
1435                         break;
1436                 }
1437         }
1438
1439         uint32_t advanced;
1440
1441         if(!(hdr.ctl & ACK)) {
1442                 advanced = 0;
1443                 goto skip_ack;
1444         }
1445
1446         // 3. Advance snd.una
1447
1448         advanced = seqdiff(hdr.ack, c->snd.una);
1449
1450         if(advanced) {
1451                 // RTT measurement
1452                 if(c->rtt_start.tv_sec) {
1453                         if(c->rtt_seq == hdr.ack) {
1454                                 struct timespec now;
1455                                 clock_gettime(UTCP_CLOCK, &now);
1456                                 int32_t diff = timespec_diff_usec(&now, &c->rtt_start);
1457                                 update_rtt(c, diff);
1458                                 c->rtt_start.tv_sec = 0;
1459                         } else if(c->rtt_seq < hdr.ack) {
1460                                 debug(c, "cancelling RTT measurement: %u < %u\n", c->rtt_seq, hdr.ack);
1461                                 c->rtt_start.tv_sec = 0;
1462                         }
1463                 }
1464
1465                 int32_t data_acked = advanced;
1466
1467                 switch(c->state) {
1468                 case SYN_SENT:
1469                 case SYN_RECEIVED:
1470                         data_acked--;
1471                         break;
1472
1473                 // TODO: handle FIN as well.
1474                 default:
1475                         break;
1476                 }
1477
1478                 assert(data_acked >= 0);
1479
1480 #ifndef NDEBUG
1481                 int32_t bufused = seqdiff(c->snd.last, c->snd.una);
1482                 assert(data_acked <= bufused);
1483 #endif
1484
1485                 if(data_acked) {
1486                         buffer_discard(&c->sndbuf, data_acked);
1487                 }
1488
1489                 // Also advance snd.nxt if possible
1490                 if(seqdiff(c->snd.nxt, hdr.ack) < 0) {
1491                         c->snd.nxt = hdr.ack;
1492                 }
1493
1494                 c->snd.una = hdr.ack;
1495
1496                 if(c->dupack) {
1497                         if(c->dupack >= 3) {
1498                                 debug(c, "fast recovery ended\n");
1499                                 c->snd.cwnd = c->snd.ssthresh;
1500                         }
1501
1502                         c->dupack = 0;
1503                 }
1504
1505                 // Increase the congestion window according to RFC 5681
1506                 if(c->snd.cwnd < c->snd.ssthresh) {
1507                         c->snd.cwnd += min(advanced, utcp->mss); // eq. 2
1508                 } else {
1509                         c->snd.cwnd += max(1, (utcp->mss * utcp->mss) / c->snd.cwnd); // eq. 3
1510                 }
1511
1512                 if(c->snd.cwnd > c->sndbuf.maxsize) {
1513                         c->snd.cwnd = c->sndbuf.maxsize;
1514                 }
1515
1516                 debug_cwnd(c);
1517
1518                 // Check if we have sent a FIN that is now ACKed.
1519                 switch(c->state) {
1520                 case FIN_WAIT_1:
1521                         if(c->snd.una == c->snd.last) {
1522                                 set_state(c, FIN_WAIT_2);
1523                         }
1524
1525                         break;
1526
1527                 case CLOSING:
1528                         if(c->snd.una == c->snd.last) {
1529                                 clock_gettime(UTCP_CLOCK, &c->conn_timeout);
1530                                 c->conn_timeout.tv_sec += utcp->timeout;
1531                                 set_state(c, TIME_WAIT);
1532                         }
1533
1534                         break;
1535
1536                 default:
1537                         break;
1538                 }
1539         } else {
1540                 if(!len && is_reliable(c) && c->snd.una != c->snd.last) {
1541                         c->dupack++;
1542                         debug(c, "duplicate ACK %d\n", c->dupack);
1543
1544                         if(c->dupack == 3) {
1545                                 // RFC 5681 fast recovery
1546                                 debug(c, "fast recovery started\n", c->dupack);
1547                                 uint32_t flightsize = seqdiff(c->snd.nxt, c->snd.una);
1548                                 c->snd.ssthresh = max(flightsize / 2, utcp->mss * 2); // eq. 4
1549                                 c->snd.cwnd = min(c->snd.ssthresh + 3 * utcp->mss, c->sndbuf.maxsize);
1550
1551                                 if(c->snd.cwnd > c->sndbuf.maxsize) {
1552                                         c->snd.cwnd = c->sndbuf.maxsize;
1553                                 }
1554
1555                                 debug_cwnd(c);
1556
1557                                 fast_retransmit(c);
1558                         } else if(c->dupack > 3) {
1559                                 c->snd.cwnd += utcp->mss;
1560
1561                                 if(c->snd.cwnd > c->sndbuf.maxsize) {
1562                                         c->snd.cwnd = c->sndbuf.maxsize;
1563                                 }
1564
1565                                 debug_cwnd(c);
1566                         }
1567
1568                         // We got an ACK which indicates the other side did get one of our packets.
1569                         // Reset the retransmission timer to avoid going to slow start,
1570                         // but don't touch the connection timeout.
1571                         start_retransmit_timer(c);
1572                 }
1573         }
1574
1575         // 4. Update timers
1576
1577         if(advanced) {
1578                 if(c->snd.una == c->snd.last) {
1579                         stop_retransmit_timer(c);
1580                         timespec_clear(&c->conn_timeout);
1581                 } else if(is_reliable(c)) {
1582                         start_retransmit_timer(c);
1583                         clock_gettime(UTCP_CLOCK, &c->conn_timeout);
1584                         c->conn_timeout.tv_sec += utcp->timeout;
1585                 }
1586         }
1587
1588 skip_ack:
1589         // 5. Process SYN stuff
1590
1591         if(hdr.ctl & SYN) {
1592                 switch(c->state) {
1593                 case SYN_SENT:
1594
1595                         // This is a SYNACK. It should always have ACKed the SYN.
1596                         if(!advanced) {
1597                                 goto reset;
1598                         }
1599
1600                         c->rcv.irs = hdr.seq;
1601                         c->rcv.nxt = hdr.seq;
1602
1603                         if(c->shut_wr) {
1604                                 c->snd.last++;
1605                                 set_state(c, FIN_WAIT_1);
1606                         } else {
1607                                 set_state(c, ESTABLISHED);
1608                         }
1609
1610                         // TODO: notify application of this somehow.
1611                         break;
1612
1613                 case SYN_RECEIVED:
1614                         // This is a retransmit of a SYN, send back the SYNACK.
1615                         goto synack;
1616
1617                 case ESTABLISHED:
1618                 case FIN_WAIT_1:
1619                 case FIN_WAIT_2:
1620                 case CLOSE_WAIT:
1621                 case CLOSING:
1622                 case LAST_ACK:
1623                 case TIME_WAIT:
1624                         // Ehm, no. We should never receive a second SYN.
1625                         return 0;
1626
1627                 default:
1628 #ifdef UTCP_DEBUG
1629                         abort();
1630 #endif
1631                         return 0;
1632                 }
1633
1634                 // SYN counts as one sequence number
1635                 c->rcv.nxt++;
1636         }
1637
1638         // 6. Process new data
1639
1640         if(c->state == SYN_RECEIVED) {
1641                 // This is the ACK after the SYNACK. It should always have ACKed the SYNACK.
1642                 if(!advanced) {
1643                         goto reset;
1644                 }
1645
1646                 // Are we still LISTENing?
1647                 if(utcp->accept) {
1648                         utcp->accept(c, c->src);
1649                 }
1650
1651                 if(c->state != ESTABLISHED) {
1652                         set_state(c, CLOSED);
1653                         c->reapable = true;
1654                         goto reset;
1655                 }
1656         }
1657
1658         if(len) {
1659                 switch(c->state) {
1660                 case SYN_SENT:
1661                 case SYN_RECEIVED:
1662                         // This should never happen.
1663 #ifdef UTCP_DEBUG
1664                         abort();
1665 #endif
1666                         return 0;
1667
1668                 case ESTABLISHED:
1669                 case FIN_WAIT_1:
1670                 case FIN_WAIT_2:
1671                         break;
1672
1673                 case CLOSE_WAIT:
1674                 case CLOSING:
1675                 case LAST_ACK:
1676                 case TIME_WAIT:
1677                         // Ehm no, We should never receive more data after a FIN.
1678                         goto reset;
1679
1680                 default:
1681 #ifdef UTCP_DEBUG
1682                         abort();
1683 #endif
1684                         return 0;
1685                 }
1686
1687                 handle_incoming_data(c, hdr.seq, ptr, len);
1688         }
1689
1690         // 7. Process FIN stuff
1691
1692         if((hdr.ctl & FIN) && (!is_reliable(c) || hdr.seq + len == c->rcv.nxt)) {
1693                 switch(c->state) {
1694                 case SYN_SENT:
1695                 case SYN_RECEIVED:
1696                         // This should never happen.
1697 #ifdef UTCP_DEBUG
1698                         abort();
1699 #endif
1700                         break;
1701
1702                 case ESTABLISHED:
1703                         set_state(c, CLOSE_WAIT);
1704                         break;
1705
1706                 case FIN_WAIT_1:
1707                         set_state(c, CLOSING);
1708                         break;
1709
1710                 case FIN_WAIT_2:
1711                         clock_gettime(UTCP_CLOCK, &c->conn_timeout);
1712                         c->conn_timeout.tv_sec += utcp->timeout;
1713                         set_state(c, TIME_WAIT);
1714                         break;
1715
1716                 case CLOSE_WAIT:
1717                 case CLOSING:
1718                 case LAST_ACK:
1719                 case TIME_WAIT:
1720                         // Ehm, no. We should never receive a second FIN.
1721                         goto reset;
1722
1723                 default:
1724 #ifdef UTCP_DEBUG
1725                         abort();
1726 #endif
1727                         break;
1728                 }
1729
1730                 // FIN counts as one sequence number
1731                 c->rcv.nxt++;
1732                 len++;
1733
1734                 // Inform the application that the peer closed its end of the connection.
1735                 if(c->recv) {
1736                         errno = 0;
1737                         c->recv(c, NULL, 0);
1738                 }
1739         }
1740
1741         // Now we send something back if:
1742         // - we received data, so we have to send back an ACK
1743         //   -> sendatleastone = true
1744         // - or we got an ack, so we should maybe send a bit more data
1745         //   -> sendatleastone = false
1746
1747         if(is_reliable(c) || hdr.ctl & SYN || hdr.ctl & FIN) {
1748                 ack(c, has_data);
1749         }
1750
1751         return 0;
1752
1753 reset:
1754         swap_ports(&hdr);
1755         hdr.wnd = 0;
1756         hdr.aux = 0;
1757
1758         if(hdr.ctl & ACK) {
1759                 hdr.seq = hdr.ack;
1760                 hdr.ctl = RST;
1761         } else {
1762                 hdr.ack = hdr.seq + len;
1763                 hdr.seq = 0;
1764                 hdr.ctl = RST | ACK;
1765         }
1766
1767         print_packet(c, "send", &hdr, sizeof(hdr));
1768         utcp->send(utcp, &hdr, sizeof(hdr));
1769         return 0;
1770
1771 }
1772
1773 int utcp_shutdown(struct utcp_connection *c, int dir) {
1774         debug(c, "shutdown %d at %u\n", dir, c ? c->snd.last : 0);
1775
1776         if(!c) {
1777                 errno = EFAULT;
1778                 return -1;
1779         }
1780
1781         if(c->reapable) {
1782                 debug(c, "shutdown() called on closed connection\n");
1783                 errno = EBADF;
1784                 return -1;
1785         }
1786
1787         if(!(dir == UTCP_SHUT_RD || dir == UTCP_SHUT_WR || dir == UTCP_SHUT_RDWR)) {
1788                 errno = EINVAL;
1789                 return -1;
1790         }
1791
1792         // TCP does not have a provision for stopping incoming packets.
1793         // The best we can do is to just ignore them.
1794         if(dir == UTCP_SHUT_RD || dir == UTCP_SHUT_RDWR) {
1795                 c->recv = NULL;
1796         }
1797
1798         // The rest of the code deals with shutting down writes.
1799         if(dir == UTCP_SHUT_RD) {
1800                 return 0;
1801         }
1802
1803         // Only process shutting down writes once.
1804         if(c->shut_wr) {
1805                 return 0;
1806         }
1807
1808         c->shut_wr = true;
1809
1810         switch(c->state) {
1811         case CLOSED:
1812         case LISTEN:
1813                 errno = ENOTCONN;
1814                 return -1;
1815
1816         case SYN_SENT:
1817                 return 0;
1818
1819         case SYN_RECEIVED:
1820         case ESTABLISHED:
1821                 set_state(c, FIN_WAIT_1);
1822                 break;
1823
1824         case FIN_WAIT_1:
1825         case FIN_WAIT_2:
1826                 return 0;
1827
1828         case CLOSE_WAIT:
1829                 set_state(c, CLOSING);
1830                 break;
1831
1832         case CLOSING:
1833         case LAST_ACK:
1834         case TIME_WAIT:
1835                 return 0;
1836         }
1837
1838         c->snd.last++;
1839
1840         ack(c, false);
1841
1842         if(!timespec_isset(&c->rtrx_timeout)) {
1843                 start_retransmit_timer(c);
1844         }
1845
1846         return 0;
1847 }
1848
1849 static bool reset_connection(struct utcp_connection *c) {
1850         if(!c) {
1851                 errno = EFAULT;
1852                 return false;
1853         }
1854
1855         if(c->reapable) {
1856                 debug(c, "abort() called on closed connection\n");
1857                 errno = EBADF;
1858                 return false;
1859         }
1860
1861         c->recv = NULL;
1862         c->poll = NULL;
1863
1864         switch(c->state) {
1865         case CLOSED:
1866                 return true;
1867
1868         case LISTEN:
1869         case SYN_SENT:
1870         case CLOSING:
1871         case LAST_ACK:
1872         case TIME_WAIT:
1873                 set_state(c, CLOSED);
1874                 return true;
1875
1876         case SYN_RECEIVED:
1877         case ESTABLISHED:
1878         case FIN_WAIT_1:
1879         case FIN_WAIT_2:
1880         case CLOSE_WAIT:
1881                 set_state(c, CLOSED);
1882                 break;
1883         }
1884
1885         // Send RST
1886
1887         struct hdr hdr;
1888
1889         hdr.src = c->src;
1890         hdr.dst = c->dst;
1891         hdr.seq = c->snd.nxt;
1892         hdr.ack = 0;
1893         hdr.wnd = 0;
1894         hdr.ctl = RST;
1895
1896         print_packet(c, "send", &hdr, sizeof(hdr));
1897         c->utcp->send(c->utcp, &hdr, sizeof(hdr));
1898         return true;
1899 }
1900
1901 // Closes all the opened connections
1902 void utcp_abort_all_connections(struct utcp *utcp) {
1903         if(!utcp) {
1904                 errno = EINVAL;
1905                 return;
1906         }
1907
1908         for(int i = 0; i < utcp->nconnections; i++) {
1909                 struct utcp_connection *c = utcp->connections[i];
1910
1911                 if(c->reapable || c->state == CLOSED) {
1912                         continue;
1913                 }
1914
1915                 utcp_recv_t old_recv = c->recv;
1916                 utcp_poll_t old_poll = c->poll;
1917
1918                 reset_connection(c);
1919
1920                 if(old_recv) {
1921                         errno = 0;
1922                         old_recv(c, NULL, 0);
1923                 }
1924
1925                 if(old_poll && !c->reapable) {
1926                         errno = 0;
1927                         old_poll(c, 0);
1928                 }
1929         }
1930
1931         return;
1932 }
1933
1934 int utcp_close(struct utcp_connection *c) {
1935         if(utcp_shutdown(c, SHUT_RDWR) && errno != ENOTCONN) {
1936                 return -1;
1937         }
1938
1939         c->recv = NULL;
1940         c->poll = NULL;
1941         c->reapable = true;
1942         return 0;
1943 }
1944
1945 int utcp_abort(struct utcp_connection *c) {
1946         if(!reset_connection(c)) {
1947                 return -1;
1948         }
1949
1950         c->reapable = true;
1951         return 0;
1952 }
1953
1954 /* Handle timeouts.
1955  * One call to this function will loop through all connections,
1956  * checking if something needs to be resent or not.
1957  * The return value is the time to the next timeout in milliseconds,
1958  * or maybe a negative value if the timeout is infinite.
1959  */
1960 struct timespec utcp_timeout(struct utcp *utcp) {
1961         struct timespec now;
1962         clock_gettime(UTCP_CLOCK, &now);
1963         struct timespec next = {now.tv_sec + 3600, now.tv_nsec};
1964
1965         for(int i = 0; i < utcp->nconnections; i++) {
1966                 struct utcp_connection *c = utcp->connections[i];
1967
1968                 if(!c) {
1969                         continue;
1970                 }
1971
1972                 // delete connections that have been utcp_close()d.
1973                 if(c->state == CLOSED) {
1974                         if(c->reapable) {
1975                                 debug(c, "reaping\n");
1976                                 free_connection(c);
1977                                 i--;
1978                         }
1979
1980                         continue;
1981                 }
1982
1983                 if(timespec_isset(&c->conn_timeout) && timespec_lt(&c->conn_timeout, &now)) {
1984                         errno = ETIMEDOUT;
1985                         c->state = CLOSED;
1986
1987                         if(c->recv) {
1988                                 c->recv(c, NULL, 0);
1989                         }
1990
1991                         if(c->poll && !c->reapable) {
1992                                 c->poll(c, 0);
1993                         }
1994
1995                         continue;
1996                 }
1997
1998                 if(timespec_isset(&c->rtrx_timeout) && timespec_lt(&c->rtrx_timeout, &now)) {
1999                         debug(c, "retransmitting after timeout\n");
2000                         retransmit(c);
2001                 }
2002
2003                 if(c->poll) {
2004                         if((c->state == ESTABLISHED || c->state == CLOSE_WAIT)) {
2005                                 uint32_t len =  buffer_free(&c->sndbuf);
2006
2007                                 if(len) {
2008                                         c->poll(c, len);
2009                                 }
2010                         } else if(c->state == CLOSED) {
2011                                 c->poll(c, 0);
2012                         }
2013                 }
2014
2015                 if(timespec_isset(&c->conn_timeout) && timespec_lt(&c->conn_timeout, &next)) {
2016                         next = c->conn_timeout;
2017                 }
2018
2019                 if(timespec_isset(&c->rtrx_timeout) && timespec_lt(&c->rtrx_timeout, &next)) {
2020                         next = c->rtrx_timeout;
2021                 }
2022         }
2023
2024         struct timespec diff;
2025
2026         timespec_sub(&next, &now, &diff);
2027
2028         return diff;
2029 }
2030
2031 bool utcp_is_active(struct utcp *utcp) {
2032         if(!utcp) {
2033                 return false;
2034         }
2035
2036         for(int i = 0; i < utcp->nconnections; i++)
2037                 if(utcp->connections[i]->state != CLOSED && utcp->connections[i]->state != TIME_WAIT) {
2038                         return true;
2039                 }
2040
2041         return false;
2042 }
2043
2044 struct utcp *utcp_init(utcp_accept_t accept, utcp_pre_accept_t pre_accept, utcp_send_t send, void *priv) {
2045         if(!send) {
2046                 errno = EFAULT;
2047                 return NULL;
2048         }
2049
2050         struct utcp *utcp = calloc(1, sizeof(*utcp));
2051
2052         if(!utcp) {
2053                 return NULL;
2054         }
2055
2056         if(!CLOCK_GRANULARITY) {
2057                 struct timespec res;
2058                 clock_getres(UTCP_CLOCK, &res);
2059                 CLOCK_GRANULARITY = res.tv_sec * USEC_PER_SEC + res.tv_nsec / 1000;
2060         }
2061
2062         utcp->accept = accept;
2063         utcp->pre_accept = pre_accept;
2064         utcp->send = send;
2065         utcp->priv = priv;
2066         utcp_set_mtu(utcp, DEFAULT_MTU);
2067         utcp->timeout = DEFAULT_USER_TIMEOUT; // sec
2068         utcp->rto = START_RTO; // usec
2069
2070         return utcp;
2071 }
2072
2073 void utcp_exit(struct utcp *utcp) {
2074         if(!utcp) {
2075                 return;
2076         }
2077
2078         for(int i = 0; i < utcp->nconnections; i++) {
2079                 struct utcp_connection *c = utcp->connections[i];
2080
2081                 if(!c->reapable) {
2082                         if(c->recv) {
2083                                 c->recv(c, NULL, 0);
2084                         }
2085
2086                         if(c->poll && !c->reapable) {
2087                                 c->poll(c, 0);
2088                         }
2089                 }
2090
2091                 buffer_exit(&c->rcvbuf);
2092                 buffer_exit(&c->sndbuf);
2093                 free(c);
2094         }
2095
2096         free(utcp->connections);
2097         free(utcp);
2098 }
2099
2100 uint16_t utcp_get_mtu(struct utcp *utcp) {
2101         return utcp ? utcp->mtu : 0;
2102 }
2103
2104 uint16_t utcp_get_mss(struct utcp *utcp) {
2105         return utcp ? utcp->mss : 0;
2106 }
2107
2108 void utcp_set_mtu(struct utcp *utcp, uint16_t mtu) {
2109         if(!utcp) {
2110                 return;
2111         }
2112
2113         if(mtu <= sizeof(struct hdr)) {
2114                 return;
2115         }
2116
2117         if(mtu > utcp->mtu) {
2118                 char *new = realloc(utcp->pkt, mtu + sizeof(struct hdr));
2119
2120                 if(!new) {
2121                         return;
2122                 }
2123
2124                 utcp->pkt = new;
2125         }
2126
2127         utcp->mtu = mtu;
2128         utcp->mss = mtu - sizeof(struct hdr);
2129 }
2130
2131 void utcp_reset_timers(struct utcp *utcp) {
2132         if(!utcp) {
2133                 return;
2134         }
2135
2136         struct timespec now, then;
2137
2138         clock_gettime(UTCP_CLOCK, &now);
2139
2140         then = now;
2141
2142         then.tv_sec += utcp->timeout;
2143
2144         for(int i = 0; i < utcp->nconnections; i++) {
2145                 struct utcp_connection *c = utcp->connections[i];
2146
2147                 if(c->reapable) {
2148                         continue;
2149                 }
2150
2151                 if(timespec_isset(&c->rtrx_timeout)) {
2152                         c->rtrx_timeout = now;
2153                 }
2154
2155                 if(timespec_isset(&c->conn_timeout)) {
2156                         c->conn_timeout = then;
2157                 }
2158
2159                 c->rtt_start.tv_sec = 0;
2160         }
2161
2162         if(utcp->rto > START_RTO) {
2163                 utcp->rto = START_RTO;
2164         }
2165 }
2166
2167 int utcp_get_user_timeout(struct utcp *u) {
2168         return u ? u->timeout : 0;
2169 }
2170
2171 void utcp_set_user_timeout(struct utcp *u, int timeout) {
2172         if(u) {
2173                 u->timeout = timeout;
2174         }
2175 }
2176
2177 size_t utcp_get_sndbuf(struct utcp_connection *c) {
2178         return c ? c->sndbuf.maxsize : 0;
2179 }
2180
2181 size_t utcp_get_sndbuf_free(struct utcp_connection *c) {
2182         if(!c) {
2183                 return 0;
2184         }
2185
2186         switch(c->state) {
2187         case SYN_SENT:
2188         case SYN_RECEIVED:
2189         case ESTABLISHED:
2190         case CLOSE_WAIT:
2191                 return buffer_free(&c->sndbuf);
2192
2193         default:
2194                 return 0;
2195         }
2196 }
2197
2198 void utcp_set_sndbuf(struct utcp_connection *c, size_t size) {
2199         if(!c) {
2200                 return;
2201         }
2202
2203         c->sndbuf.maxsize = size;
2204
2205         if(c->sndbuf.maxsize != size) {
2206                 c->sndbuf.maxsize = -1;
2207         }
2208 }
2209
2210 size_t utcp_get_rcvbuf(struct utcp_connection *c) {
2211         return c ? c->rcvbuf.maxsize : 0;
2212 }
2213
2214 size_t utcp_get_rcvbuf_free(struct utcp_connection *c) {
2215         if(c && (c->state == ESTABLISHED || c->state == CLOSE_WAIT)) {
2216                 return buffer_free(&c->rcvbuf);
2217         } else {
2218                 return 0;
2219         }
2220 }
2221
2222 void utcp_set_rcvbuf(struct utcp_connection *c, size_t size) {
2223         if(!c) {
2224                 return;
2225         }
2226
2227         c->rcvbuf.maxsize = size;
2228
2229         if(c->rcvbuf.maxsize != size) {
2230                 c->rcvbuf.maxsize = -1;
2231         }
2232 }
2233
2234 size_t utcp_get_sendq(struct utcp_connection *c) {
2235         return c->sndbuf.used;
2236 }
2237
2238 size_t utcp_get_recvq(struct utcp_connection *c) {
2239         return c->rcvbuf.used;
2240 }
2241
2242 bool utcp_get_nodelay(struct utcp_connection *c) {
2243         return c ? c->nodelay : false;
2244 }
2245
2246 void utcp_set_nodelay(struct utcp_connection *c, bool nodelay) {
2247         if(c) {
2248                 c->nodelay = nodelay;
2249         }
2250 }
2251
2252 bool utcp_get_keepalive(struct utcp_connection *c) {
2253         return c ? c->keepalive : false;
2254 }
2255
2256 void utcp_set_keepalive(struct utcp_connection *c, bool keepalive) {
2257         if(c) {
2258                 c->keepalive = keepalive;
2259         }
2260 }
2261
2262 size_t utcp_get_outq(struct utcp_connection *c) {
2263         return c ? seqdiff(c->snd.nxt, c->snd.una) : 0;
2264 }
2265
2266 void utcp_set_recv_cb(struct utcp_connection *c, utcp_recv_t recv) {
2267         if(c) {
2268                 c->recv = recv;
2269         }
2270 }
2271
2272 void utcp_set_poll_cb(struct utcp_connection *c, utcp_poll_t poll) {
2273         if(c) {
2274                 c->poll = poll;
2275         }
2276 }
2277
2278 void utcp_set_accept_cb(struct utcp *utcp, utcp_accept_t accept, utcp_pre_accept_t pre_accept) {
2279         if(utcp) {
2280                 utcp->accept = accept;
2281                 utcp->pre_accept = pre_accept;
2282         }
2283 }
2284
2285 void utcp_expect_data(struct utcp_connection *c, bool expect) {
2286         if(!c || c->reapable) {
2287                 return;
2288         }
2289
2290         if(!(c->state == ESTABLISHED || c->state == FIN_WAIT_1 || c->state == FIN_WAIT_2)) {
2291                 return;
2292         }
2293
2294         if(expect) {
2295                 // If we expect data, start the connection timer.
2296                 if(!timespec_isset(&c->conn_timeout)) {
2297                         clock_gettime(UTCP_CLOCK, &c->conn_timeout);
2298                         c->conn_timeout.tv_sec += c->utcp->timeout;
2299                 }
2300         } else {
2301                 // If we want to cancel expecting data, only clear the timer when there is no unACKed data.
2302                 if(c->snd.una == c->snd.last) {
2303                         timespec_clear(&c->conn_timeout);
2304                 }
2305         }
2306 }
2307
2308 void utcp_offline(struct utcp *utcp, bool offline) {
2309         struct timespec now;
2310         clock_gettime(UTCP_CLOCK, &now);
2311
2312         for(int i = 0; i < utcp->nconnections; i++) {
2313                 struct utcp_connection *c = utcp->connections[i];
2314
2315                 if(c->reapable) {
2316                         continue;
2317                 }
2318
2319                 utcp_expect_data(c, offline);
2320
2321                 if(!offline) {
2322                         if(timespec_isset(&c->rtrx_timeout)) {
2323                                 c->rtrx_timeout = now;
2324                         }
2325
2326                         utcp->connections[i]->rtt_start.tv_sec = 0;
2327                 }
2328         }
2329
2330         if(!offline && utcp->rto > START_RTO) {
2331                 utcp->rto = START_RTO;
2332         }
2333 }
2334
2335 void utcp_set_clock_granularity(long granularity) {
2336         CLOCK_GRANULARITY = granularity;
2337 }