/usr/include/itpp/protocol/tcp.h is in libitpp-dev 4.3.1-6.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
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 | /*!
* \file
* \brief Definition of Transport Control Protocol (TCP)
* \author Krister Norlund
*
* -------------------------------------------------------------------------
*
* Copyright (C) 1995-2010 (see AUTHORS file for a list of contributors)
*
* This file is part of IT++ - a C++ library of mathematical, signal
* processing, speech processing, and communications classes and functions.
*
* IT++ is free software: you can redistribute it and/or modify it under the
* terms of the GNU General Public License as published by the Free Software
* Foundation, either version 3 of the License, or (at your option) any
* later version.
*
* IT++ is distributed in the hope that it will be useful, but WITHOUT ANY
* WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
* FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
* details.
*
* You should have received a copy of the GNU General Public License along
* with IT++. If not, see <http://www.gnu.org/licenses/>.
*
* -------------------------------------------------------------------------
*
* Copyright (c) 2000-2004 IKR (formerly IND), University of Stuttgart
* This file is part of the IKR (formerly IND) TCP Library.
*/
#ifndef TCP_H
#define TCP_H
#include <itpp/base/vec.h>
#if (defined(_MSC_VER) && defined(ITPP_SHARED_LIB) && !(defined(itpp_EXPORTS) || defined(itpp_debug_EXPORTS)))
#ifndef ITPP_PROTOCOL_EXCLUDED
#define ITPP_PROTOCOL_EXCLUDED
#pragma message( "PROTOCOL definitions are not available for MSVC shared builds" )
#endif
#else
#include <itpp/base/converters.h>
#include <itpp/protocol/packet.h>
#include <itpp/protocol/events.h>
namespace itpp
{
//! \addtogroup protocol
//@{
/*!
TSequenceNumber represents a sequence number in a byte stream. As
simulations may get quite long sequence numbers based on 32 bit integers
may overflow. This is not a major problem as long as this overflow
is considered when comparing sequence numbers. TSequenceNumber defines
relational operators base on subtraction instead of using built-in
int relational operators. This approach works well for arbitrary long
simulations as long as the "real" sequence numbers (i.e. the numbers
seen without overflow) compared to each other do not differ by more than
2^31-1 (INT_MAX) which is not a serious restriction in a realistic
TCP scenario.<p>
@sa TTCPSegment
@author Bodamer
*/
class Sequence_Number
{
public:
//! Default constructor
Sequence_Number() : seq(0) { }
//! ADD DOCUMENTATION HERE
Sequence_Number(const Sequence_Number &n) : seq(n.seq) { }
//! ADD DOCUMENTATION HERE
Sequence_Number &operator=(const Sequence_Number &n) { seq = n.seq; return *this; }
//! ADD DOCUMENTATION HERE
Sequence_Number &operator=(const int &rep) { seq = rep; return *this; }
//relational operators
//! ADD DOCUMENTATION HERE
bool operator==(const Sequence_Number &n) const { return seq == n.seq; }
//! ADD DOCUMENTATION HERE
bool operator!=(const Sequence_Number &n) const { return seq != n.seq; }
//! ADD DOCUMENTATION HERE
bool operator>(const Sequence_Number &n) const { return (seq - n.seq) > 0; }
//! ADD DOCUMENTATION HERE
bool operator>=(const Sequence_Number &n) const { return (seq - n.seq) >= 0; }
//! ADD DOCUMENTATION HERE
bool operator<(const Sequence_Number &n) const { return (seq - n.seq) < 0; }
//! ADD DOCUMENTATION HERE
bool operator<=(const Sequence_Number &n) const { return (seq - n.seq) <= 0; }
//addition and subtraction
//! ADD DOCUMENTATION HERE
Sequence_Number operator+(const int n) const { return Sequence_Number(seq + n); }
//! ADD DOCUMENTATION HERE
Sequence_Number &operator+=(const int n) { seq += n; return *this; }
//! ADD DOCUMENTATION HERE
Sequence_Number operator-(const int n) const { return Sequence_Number(seq - n); }
//! ADD DOCUMENTATION HERE
Sequence_Number &operator-=(const int n) { seq -= n; return *this; }
//! ADD DOCUMENTATION HERE
int operator-(const Sequence_Number &n) const { return seq - n.seq; }
//! Access to internal representation
int value() const { return seq; }
//! ADD DOCUMENTATION HERE
friend Sequence_Number operator+(const int n1, const Sequence_Number &n2) { return Sequence_Number(n1 + n2.seq); }
//! ADD DOCUMENTATION HERE
friend std::ostream &operator<<(std::ostream &os, const Sequence_Number &n) { os << n.seq; return os; }
protected:
//! ADD DOCUMENTATION HERE
Sequence_Number(int n) : seq(n) {}
//! ADD DOCUMENTATION HERE
int seq;
};
//! ADD DOCUMENTATION HERE
inline const Sequence_Number & min(const Sequence_Number &n1, const Sequence_Number &n2) { return (n1 < n2) ? n1 : n2; }
//! ADD DOCUMENTATION HERE
inline const Sequence_Number & max(const Sequence_Number &n1, const Sequence_Number &n2) { return (n1 > n2) ? n1 : n2; }
/*!
TCP is a byte oriented protcol. Parts of the byte stream that is
transmitted are called segments. They are identified by sequence numbers
TCP_Segment contains fileds for the beginning sequence number and the
sequence number of the first byte in the following segment (i.e. the
sequence number of the last byte + 1). TCP_Segment provides several
const methods to compare segments and to check whether they overlap.
TCP_Segment is, e.g., used in TTCPPacket and TTCPReceiverBuffer.<p>
@sa Sequence_Number
@sa TTCPPacket
@sa TTCPReceiverBuffer
@author Lorang
@author Bodamer
*/
class TCP_Segment
{
public:
//! ADD DOCUMENTATION HERE
TCP_Segment();
//! ADD DOCUMENTATION HERE
TCP_Segment(const Sequence_Number &sn_begin, const Sequence_Number &sn_end);
//! ADD DOCUMENTATION HERE
TCP_Segment(const TCP_Segment &segment);
//modification
//! ADD DOCUMENTATION HERE
TCP_Segment &operator=(const TCP_Segment &segment);
//! ADD DOCUMENTATION HERE
void set_begin(const Sequence_Number &sn);
//! ADD DOCUMENTATION HERE
void set_end(const Sequence_Number &sn);
//! ADD DOCUMENTATION HERE
void combine(const TCP_Segment &segment);
//query
//! ADD DOCUMENTATION HERE
bool operator==(const TCP_Segment &segment) const;
//! ADD DOCUMENTATION HERE
bool operator!=(const TCP_Segment &segment) const;
//! ADD DOCUMENTATION HERE
bool can_be_combined(const TCP_Segment &segment) const;
//! ADD DOCUMENTATION HERE
bool is_contained(const TCP_Segment &segment) const;
//! ADD DOCUMENTATION HERE
unsigned length() const;
//! ADD DOCUMENTATION HERE
Sequence_Number begin() const { return seq_begin; }
//! ADD DOCUMENTATION HERE
Sequence_Number end() const { return seq_end; }
//! ADD DOCUMENTATION HERE
friend std::ostream & operator<<(std::ostream &os, const TCP_Segment &segment);
protected:
Sequence_Number seq_begin; /**< \brief no. of first byte of segment */
Sequence_Number seq_end; /**< \brief no. of last byte of segment + 1 */
};
/*!
TCP_Packet is an IP packet with additional TCP header fields.
Messages of this type are used for communication between TCP sender
and receiver (data and ACKs).<p>
The class contains a segment of type TCP_Segment with sequence numbers
for begin and end of the segment. This is only used in data packets
(i.e. from sender to receiver) while there are also fields containing
sequence number for the next expected segment (used to ACK data
packets, i.e. from receiver to sender) and for the advertised
receiver window.<p>
TCP_Packet has an additional field for the session id which my be compared
with the pair (source port, destination port) in a real world
TCP/IP packet.<p>
Furthermore, some debug info containing state information for TCP sender
or receiver may be attached to the message.<p>
@sa TCP_Segment
@sa TCP_Packet::TDebugInfo
@author Grevent
@author Lorang
@author Bodamer
*/
class TCP_Packet : public itpp::Packet
{
public:
//! ADD DOCUMENTATION HERE
TCP_Packet();
//! ADD DOCUMENTATION HERE
TCP_Packet(const TCP_Packet &packet);
//! ADD DOCUMENTATION HERE
virtual ~TCP_Packet();
//! ADD DOCUMENTATION HERE
virtual TCP_Packet &clone() const;
//TCP window mechanism
//! ADD DOCUMENTATION HERE
void set_segment(const TCP_Segment &seg) { fSegment = seg; }
//! ADD DOCUMENTATION HERE
TCP_Segment get_segment() const { return fSegment; }
//! ADD DOCUMENTATION HERE
void set_wnd(unsigned val) { fWnd = val; }
//! ADD DOCUMENTATION HERE
unsigned get_wnd() const { return fWnd; }
//! ADD DOCUMENTATION HERE
void set_ACK(Sequence_Number val) { fACK = val; }
//! ADD DOCUMENTATION HERE
Sequence_Number get_ACK() const { return fACK; }
//session control
//! ADD DOCUMENTATION HERE
void set_session_id(int val) { fSessionId = val; }
//! ADD DOCUMENTATION HERE
int get_session_id() const { return fSessionId; }
//debugging support
//! ADD DOCUMENTATION HERE
void set_destination_port(unsigned val) { fDestinationPort = val; }
//! ADD DOCUMENTATION HERE
unsigned get_destination_port() const { return fDestinationPort; }
//! ADD DOCUMENTATION HERE
void set_source_port(unsigned val) { fSourcePort = val; }
//! ADD DOCUMENTATION HERE
unsigned get_source_port() const { return fSourcePort; }
//! ADD DOCUMENTATION HERE
void set_info(unsigned ssThresh, unsigned recWnd, unsigned cWnd, double estRTT, Sequence_Number sndUna, Sequence_Number sndNxt, bool isRtx);
//! ADD DOCUMENTATION HERE
virtual void print_header(std::ostream &) const;
protected:
//! ADD DOCUMENTATION HERE
unsigned fDestinationPort;
//! ADD DOCUMENTATION HERE
unsigned fSourcePort;
TCP_Segment fSegment; /**< \brief data segment to be transmitted */
Sequence_Number fACK; /**< \brief acknowledgment (next expected sn) */
unsigned fWnd; /**< \brief window size (advertised by receiver) */
int fSessionId; /**< \brief session identifier */
//Tracing
//! ADD DOCUMENTATION HERE
struct TDebugInfo {
unsigned fSSThresh; //!< ADD DOCUMENTATION HERE
unsigned fRecWnd; //!< ADD DOCUMENTATION HERE
unsigned fCWnd; //!< ADD DOCUMENTATION HERE
double fRTTEstimate; //!< ADD DOCUMENTATION HERE
Sequence_Number fSndUna; //!< ADD DOCUMENTATION HERE
Sequence_Number fSndNxt; //!< ADD DOCUMENTATION HERE
bool fRtxFlag; //!< ADD DOCUMENTATION HERE
};
//! ADD DOCUMENTATION HERE
TDebugInfo *fInfo;
//! ADD DOCUMENTATION HERE
friend std::ostream & operator<<(std::ostream &, TCP_Packet &);
};
/*!
TTCPSender is an entity that models TCP flow and congestion control at
the sender side. It is one of the key components of the TCP module.<p>
TTCPSender communicates with its environment via three ports: <p>
<ul>
<li>"input": receive messages from a generator/application
<li>"output": send TCP packets into the network
<li>"ackinput": receive TCP ACK packets from the network originated by
the TCP receiver
</ul><p>
The sender is activated when the user entity (e.g. a generator) offers
a message, whose length indicates the amount of data to be transferred,
or when an ACK is received from the network.
In both cases data is only sent if there is any unsent data available,
the sending window (determined by the congestion window and by the
receiver advertised window) is large enough and the silly window
syndrome avoidance algorithm is passed. Data that has been sent remains
in the sender buffer, which is modelled in a virtual manner,
until it is acknowledged. If data has been ACKed it is erased in the
sender buffer and the TCP sender tries to get new data and send it.<p>
TTCPSender can be used with different versions of congestion control:
Tahoe, Reno, and New Reno. Moreover, various options (e.g. usage
of Nagle/Karn/Go-Back-N algorithms) and parameters (e.g. mss, max cwnd,
initial values, timer granularity) can be defined in the input file.<p>
The class provides a simplified connection control via methods Setup and
Release. Those methods are only used to reset internal state variables.
No control messages are sent over the network, i.e. the sender assumes
that the corresponding methods at the receiver side are called as well.<p>
@sa TTCPSenderSet
@sa TTCPReceiver
@author Grevent
@author Lorang
@author Bodamer
*/
class TCP_Sender
{
public:
//! ADD DOCUMENTATION HERE
TCP_Sender(int label);
//! ADD DOCUMENTATION HERE
virtual ~TCP_Sender();
//connection control
//! ADD DOCUMENTATION HERE
virtual void setup();
//! ADD DOCUMENTATION HERE
virtual void release(std::string trace_filename = "");
//! Print support
virtual void print_item(std::ostream &, const std::string &);
//! ADD DOCUMENTATION HERE
virtual void set_debug(const bool enable_debug = true);
//! ADD DOCUMENTATION HERE
virtual void set_debug(bool enable_debug, bool enable_signal_debug);
//! ADD DOCUMENTATION HERE
virtual void set_trace(const bool enable_trace = true);
//! ADD DOCUMENTATION HERE
virtual void save_trace(std::string filename);
//! ADD DOCUMENTATION HERE
Signal<itpp::Packet*> tcp_send;
//! ADD DOCUMENTATION HERE
Slot<TCP_Sender, itpp::Packet*> tcp_receive_ack;
//! ADD DOCUMENTATION HERE
Slot<TCP_Sender, itpp::Packet*> tcp_socket_write;
//! ADD DOCUMENTATION HERE
Slot<TCP_Sender, std::string> tcp_release;
//Signal<itpp::Packet*> TcpSendSignal;
//Slot<TCP_Sender, itpp::Packet*> TcpRecvAckSlot;
//Slot<TCP_Sender, itpp::Packet*> SocketWriteSlot;
//Slot<TCP_Sender, string> ReleaseSlot;
private:
std::queue<itpp::Packet*> SocketWriteQueue;
virtual void InitStatistics(); /**< \brief reset statistic counters */
virtual void HandleACK(TCP_Packet &); /**< \brief process incoming ACK */
virtual void SendNewData(bool skipSWSA = false); /**< \brief send new data */
virtual void UnaRetransmit(); /**< \brief TO or fast retransmit */
virtual void FinishFastRecovery(); /**< \brief actions at end of fast recovery */
virtual void ReduceSSThresh(); /**< \brief halving on dup ACK or TO */
virtual void SendMsg(TCP_Packet & msg); /**< \brief access to network */
virtual void HandleRtxTimeout(Ttype); /**< \brief what to do after Timeout */
virtual void IdleCheck(); /**< \brief check whether SSR after idle is done */
virtual void HandleSWSATimeout(Ttype); /**< \brief handler for SWSA/Nagle timer */
virtual unsigned GetNextSegmentSize(const Sequence_Number & begin);
virtual unsigned SendWindow() const; /**< \brief min of CWnd, MaxCWnd and RecWnd */
virtual double CalcRTOValue() const; /**< \brief value for rtx timer */
virtual void SetRtxTimer();
virtual void UpdateRTTVariables(double sampleRTT); /**< \brief evaluate RTT measuremnt */
virtual void TraceCWnd();
virtual void TraceSentSeqNo(const Sequence_Number sn);
virtual void TraceACKedSeqNo(const Sequence_Number sn);
virtual void TraceRTTVariables(double sampleRTT);
virtual void TraceSSThresh();
virtual std::string GenerateFilename();
void StopTransientPhase(); /**< \brief reset statistic counters */
enum eTCPVersion {kTahoe, kReno, kNewReno};
virtual void set_label(int label);
//socket variables
unsigned fLabel; // end point identification also used at receiver
//message handling
virtual void HandleUserMessageIndication(itpp::Packet *user_data);
virtual void ReceiveMessageFromNet(itpp::Packet *msg);
//parameters parsed from input file
eTCPVersion fTCPVersion; // one of Reno, Tahoe, NewReno
unsigned fMSS; // maximum segment size
unsigned fTCPIPHeaderLength; /**< \brief additional msg length (normally 40 bytes) */
double fInitialRTT; /**< \brief RTT assumed before first measurement */
unsigned fInitialCWnd; /**< \brief initial congestion window -> RFC 2581 */
unsigned fInitialSSThresh; /**< \brief initial ssthresh -> RFC 2581 */
unsigned fMaxCWnd; /**< \brief congestion window boundary */
unsigned fDupACKThreshold; /**< \brief duplicate ACK threshold */
double fTimerGranularity; /**< \brief granularity for rtx/idle timer values */
double fMaxRTO; // max value of retransmission TO
unsigned fMaxBackoff; /**< \brief max value of backoff value (Karn) */
bool fImmediateBackoffReset; /**< \brief reset backoff on first new ACK */
bool fKarn; /**< \brief exclude rtx packets from RTTM */
bool fGoBackN; /**< \brief use go-back-N on TO rtx */
bool fFlightSizeRecovery;// use flight size on fast rec. exit
bool fRenoConservation; /**< \brief use cons. of packets on fast rec. (std.) */
bool fCarefulSSThreshReduction; /**< \brief use min(cwnd, flight size)/2 */
bool fIgnoreDupACKOnTORecovery; /**< \brief avoid fast rtx during TO recovery */
bool fCarefulMulFastRtxAvoidance; /**< \brief see RFC 2582, Section 5 */
bool fNagle; /**< \brief use Nagle algorithm */
double fSWSATimerValue; /**< \brief timer for silly wind. synd. avoidance */
bool fRestartAfterIdle; /**< \brief perform SSR after idle period */
bool fTraceCWnd; // print CWnd trace to cout
bool fTraceSentSeqNo; /**< \brief print trace of sent SNs to cout */
bool fTraceACKedSeqNo; /**< \brief print trace of received ACKs to cout */
bool fDebug; // print additional information to cout
bool fTrace; // store trace info in vectors
//session identification
int fSessionId; /**< \brief is increased when Release is called */
//TCP flow control (RFC 793)
Sequence_Number fSndUna; // lowest unacknowledged sn
Sequence_Number fSndNxt; // next byte to be sent
Sequence_Number fSndMax; /**< \brief highest byte that has been sent +1 */
unsigned fRecWnd; // receiver advertised window
unsigned fMaxRecWnd; /**< \brief maximum observed rec. window */
Sequence_Number fUserNxt; // next byte to be received from user
//TCP congestion avoidance (RFC 2581, RFC 2001, RFC 2582)
unsigned fCWnd; // congestion window
unsigned fSSThresh; /**< \brief threshold between slow start and cong. avoid. */
unsigned fDupACKCnt; /**< \brief counter for duplicate ACKs */
Sequence_Number fRecoveryDupACK; /**< \brief sndmax on 3rd dup ACK (see RFC 2582) */
Sequence_Number fRecoveryTO; /**< \brief sndmax on TO (see RFC 2582) */
//TCP timers
TTimer<TCP_Sender> fRtxTimer; /**< \brief retransmission timer */
Sequence_Number fTimUna; /**< \brief byte rtx timer is running for */
TTimer<TCP_Sender> fSWSATimer; /**< \brief SWSA/Nagle timer */
int fBackoff; /**< \brief backoff for rtx timer */
bool fPendingBackoffReset; /**< \brief reset backoff on next new ACK */
Ttype fLastSendTime; /**< \brief idle detection for SSR */
//round trip time measurement (RTTM)
double fSRTT; /**< \brief smoothed mean round trip time */
double fRTTVar; /**< \brief variance of RTT time */
double fRTTEstimate; /**< \brief created from SRTT, RTTVar and gran. */
Sequence_Number fRTTMByte; /**< \brief byte for which RTTM is running */
bool fRTTMPending; /**< \brief is currently RTTN running? */
double fRTTMStartTime; /**< \brief beginning of a measurement */
//statistic counters
unsigned long fNumberOfTimeouts;
unsigned long fNumberOfFastRetransmits;
unsigned long fNumberOfRTTMeasurements;
unsigned long fNumberOfReceivedACKs;
unsigned long fNumberOfIdleTimeouts;
vec CWnd_val;
vec CWnd_time;
int CWnd_index;
vec SSThresh_val;
vec SSThresh_time;
int SSThresh_index;
ivec sent_seq_num_val;
vec sent_seq_num_time;
int sent_seq_num_index;
ivec sender_recv_ack_seq_num_val;
vec sender_recv_ack_seq_num_time;
int sender_recv_ack_seq_num_index;
vec RTTEstimate_val;
vec RTTEstimate_time;
int RTTEstimate_index;
vec RTTsample_val;
vec RTTsample_time;
int RTTsample_index;
};
/*!
TTCPReceiverBuffer is an important part of TTCPReceiver. It is
much more complex than the buffer model at the sender side as it has to
keep track of out of order segments. The segments received are combined if
possible and the resulting non-contiguous segments are stored in a
linked list. Moreover TTCPReceiverBuffer stores the smallest sequence
number that has not (yet) been read out by the TCP receiver
("first byte").<p>
Data is written to the buffer by TTCPReceiver using method Write and
read out using method Read. Before reading data the TCP receiver has to
check whether a data block in the stream is available by calling
FirstBlockSize.<p>
Furthermore, TTCPReceiver requires information from TTCPReceiverBuffer,
e.g. about the next sequence number that is missing in the
stream (NextExpected) or the receiver window that can be advertised
to the sender (Window).<p>
@sa TTCPReceiver
@author Bodamer
@author Kutter
*/
class TCP_Receiver_Buffer
{
public:
//! ADD DOCUMENTATION HERE
TCP_Receiver_Buffer();
//! ADD DOCUMENTATION HERE
TCP_Receiver_Buffer(const TCP_Receiver_Buffer &);
//! ADD DOCUMENTATION HERE
~TCP_Receiver_Buffer();
void reset(); /**< \brief clears internal list structure */
void write(TCP_Segment newBlock); /**< \brief add segment to the queue */
void read(unsigned noOfBytes); /**< \brief read up to "noOfBytes" bytes from queue */
unsigned first_block_size() const; /**< \brief size of first complete block */
Sequence_Number first_byte() const; /**< \brief first byte stored or missing */
Sequence_Number last_byte() const; /**< \brief highest byte received (+1) */
Sequence_Number next_expected() const; /**< \brief first byte missing */
//! ADD DOCUMENTATION HERE
unsigned window() const;
std::ostream &info(std::ostream &os, int detail = 0) const; /**< \brief print info */
protected:
Sequence_Number fFirstByte; /**< \brief first byte stored or missing */
//! ADD DOCUMENTATION HERE
std::list <TCP_Segment> fBufList;
};
/*!
The TCP receiver models the receiver side of a TCP connection. It is
connected to the network via ports "input" and "ackoutput" and to an
entity modelling higher layers (e.g. a sink) via port "output".<p>
Incoming TCP messages are used to update the receiver buffer and as a
trigger to send an ACK message back to the TCP sender. If the "DelayedACK"
option is used ACKs are only sent for every second packet (unless in the
case of out of order packets).
If the received packet is not out of order (i.e. the next expected
sequence number in the byte stream has increased) new data is delivered
to the higher layer. This delivery may be delayed in the receiver itself
(user block processing delay) or by the subsequent entity if it blocks
on incoming message indications.<p>
A couple of parameters can be specified in the input file including delayed
ACK timer value and timer granularity. The value of MSS should be equal
to the one specified at the receiver side although it is only required
for delayed ACK and receiver SWSA algorithms. <p>
Like the TCP sender TTCPReceiver provides a simplified connection
control via methods Setup and Release. Those methods are only used to
reset internal state variables. No control messages are sent over
the network, i.e. the receiver assumes that the corresponding methods
at the sender side are called as well.<p>
@sa TTCPReceiverSet
@sa TTCPReceiverBuffer
@sa TCP_Sender
@author Grevent
@author Lorang
@author Bodamer
*/
class TCP_Receiver
{
public:
//! ADD DOCUMENTATION HERE
TCP_Receiver(int label);
//! ADD DOCUMENTATION HERE
virtual ~TCP_Receiver();
//name connection control
//! ADD DOCUMENTATION HERE
virtual void setup();
//! ADD DOCUMENTATION HERE
virtual void release(std::string trace_filename = "");
//message handling
itpp::Packet & get_user_message(); /**< \brief called by higher layer */
bool is_user_message_available(); /**< \brief called by higher layer */
//! ADD DOCUMENTATION HERE
virtual void set_debug(const bool enable_debug = true);
//! ADD DOCUMENTATION HERE
virtual void set_debug(bool enable_debug, bool enable_signal_debug);
//! ADD DOCUMENTATION HERE
virtual void set_trace(const bool enable_trace = true);
//! ADD DOCUMENTATION HERE
virtual void save_trace(std::string filename);
//! ADD DOCUMENTATION HERE
Signal<itpp::Packet*> tcp_send_ack;
//! ADD DOCUMENTATION HERE
Slot<TCP_Receiver, itpp::Packet*> tcp_receive;
Signal<int> tcp_new_data; /**< \brief indicate new data to higher layer */
//! ADD DOCUMENTATION HERE
Slot<TCP_Receiver, std::string> tcp_release;
private:
void IndicateUserMessage(); /**< \brief indicate new data to higher layer */
virtual void ReceiveMessageFromNet(itpp::Packet* msg); /**< \brief receive from network */
virtual void ReceiveDataPacket(TCP_Packet & packet); /**< \brief receive TCP packet */
virtual void SendACK(bool); /**< \brief send an ACK if necessary or enforced */
virtual void ScheduleACKMessage(); /**< \brief prepare ACK message for sending */
virtual void SendACKMessage(Ttype); /**< \brief called by ACK scheduling timer */
virtual void DelayedACKHandler(Ttype); /**< \brief handler for delayed ACK timer */
virtual void PeriodicACKHandler(Ttype); /**< \brief handler for periodic ACK timer */
virtual void HandleEndOfProcessing(Ttype); /**< \brief handler for user msg proc. */
virtual void TraceReceivedSeqNo(const Sequence_Number &sn);
virtual std::string GenerateFilename();
//basic variables
TCP_Receiver_Buffer fReceiverBuffer;
unsigned fLabel;
//parameters read by the parser
unsigned fTCPIPHeaderLength; /**< \brief overhead by TCP/IP headers */
unsigned fMSS; /**< \brief maximum segment size */
unsigned fBufferSize; /**< \brief size of receiver buffer */
bool fDelayedACK; /**< \brief if true, use delayed ACK */
Ttype fACKDelayTime; /**< \brief maximum time for delaying ACKs */
bool fSendPeriodicACKs; /**< \brief repeat ACKs if no data has been received */
bool fStrictPeriodicACKs; /**< \brief repeat ACKs independent of data received */
Ttype fPeriodicACKInterval;// interval after which an ACK is repeated
Ttype fACKSchedulingDelay; /**< \brief ACK delay due to proc. scheduling */
bool fACKOnBufferWrite; /**< \brief send ACK after segment has been written */
bool fACKOnBufferRead; /**< \brief send ACK when segment is read by user */
unsigned fMaxUserBlockSize; /**< \brief max size of a user read block */
unsigned fMinUserBlockSize; /**< \brief min size of a user read block */
double fUserBlockProcDelay; /**< \brief time to read data from buffer + forward */
bool fTrace; /**< \brief trace end bytes of received packets */
bool fDebug; /**< \brief print additional information */
//specific TCP variables
Sequence_Number fAdvRcvNxt; /**< \brief advertised next expected byte (ACK) */
unsigned fAdvRcvWnd; /**< \brief advertised receiver window */
//session management
int fSessionId; /**< \brief increased by 1 on connection release */
//ACK timers
TTimer<TCP_Receiver> fDelayedACKTimer; /**< \brief timer for delayed ACK */
TTimer<TCP_Receiver> fPeriodicACKTimer; /**< \brief timer for periodic ACKs */
TTimer<TCP_Receiver> fACKSchedulingTimer;
TCP_Packet * fWaitingACKMsg;
//user data delivery
Packet * fUserMessage;
TTimer<TCP_Receiver> fUserBlockProcTimer;
//statistic counters
ivec received_seq_num_val;
vec received_seq_num_time;
int received_seq_num_index;
};
// ------------------------------- Inline definitions ---------------------------------------------
inline Sequence_Number TCP_Receiver_Buffer::first_byte() const
{
return fFirstByte;
}
inline Sequence_Number TCP_Receiver_Buffer::last_byte() const
{
if (fBufList.empty()) {
return fFirstByte;
}
else {
return fBufList.back().end();
}
}
inline Sequence_Number TCP_Receiver_Buffer::next_expected() const
{
return fFirstByte + first_block_size();
}
inline void TCP_Segment::set_begin(const Sequence_Number &sn)
{
seq_begin = sn;
it_assert(seq_begin <= seq_end, "TCP_Segment::begin, end byte " + to_str(seq_end.value()) + " < begin byte " + to_str(seq_begin.value()));
}
inline void TCP_Segment::set_end(const Sequence_Number &sn)
{
seq_end = sn;
it_assert(seq_begin <= seq_end, "TCP_Segment::set_begin, end byte " + to_str(seq_end.value()) + " < begin byte " + to_str(seq_begin.value()));
}
inline bool TCP_Segment::operator==(const TCP_Segment &segment) const
{
return (this->seq_begin == segment.seq_begin) && (this->seq_end == segment.seq_end);
}
inline bool TCP_Segment::operator!=(const TCP_Segment &segment) const
{
return (this->seq_begin != segment.seq_begin) || (this->seq_end != segment.seq_end);
}
inline bool TCP_Segment::can_be_combined(const TCP_Segment &segment) const
{
return (this->seq_begin <= segment.seq_end) && (segment.seq_begin <= this->seq_end);
}
inline bool TCP_Segment::is_contained(const TCP_Segment &segment) const
{
return (segment.seq_begin <= this->seq_begin) && (this->seq_end <= segment.seq_end);
}
inline unsigned TCP_Segment::length() const
{
return seq_end - seq_begin;
}
//@}
} // namespace itpp
#endif
#endif // #ifndef TCP_H
|