Debian Codegraph @

rfc1622.html

raw
1 <pre>Network Working Group P. Francis
2 Request for Comments: 1622 NTT
3 Category: Informational May 1994
4
5
6 <span class="h1">Pip Header Processing</span>
7
8 Status of this Memo
9
10 This memo provides information for the Internet community. This memo
11 does not specify an Internet standard of any kind. Distribution of
12 this memo is unlimited.
13
14 Preamble
15
16 During 1992 and 1993, the Pip internet protocol, developed at
17 Bellcore, was one of the candidate replacments for IP. In mid 1993,
18 Pip was merged with another candidate, the Simple Internet Protocol
19 (SIP), creating SIPP (SIP Plus). While the major aspects of Pip--
20 particularly its distinction of identifier from address, and its use
21 of the source route mechanism to achieve rich routing capabilities--
22 were preserved, many of the ideas in Pip were not. The purpose of
23 this RFC and the companion RFC "Pip Near-term Architecture" are to
24 record the ideas (good and bad) of Pip.
25
26 The remainder of this document is taken verbatem from the Pip draft
27 memo of the same title that existed when the Pip project ended. As
28 such, any text that indicates that Pip is an intended replacement for
29 IP should be ignored.
30
31 Abstract
32
33 Pip is an internet protocol intended as the replacement for IP
34 version 4. Pip is a general purpose internet protocol, designed to
35 handle all forseeable internet protocol requirements. This
36 specification defines the Pip header processing for Routers and
37 Hosts.
38
39 Acknowledgements
40
41 I want to individually acknowledge Rob Coltun, Steve Deering, Ramesh
42 Govindan, Joel Halpern, John Ioannidis, Chris Petrilli, Bob Smart,
43 and Zheng Wang. I want also to acknowledge the many people from the
44 Pip working group who have participated in developing Pip. Finally,
45 I want to acknowledge the SIP protocol (or, more accurately, the
46 people behind the SIP protocol) for providing certain good ideas.
47
48
49
50
51
52 <span class="grey">Francis [Page 1]</span></pre>
53 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-2" ></span>
54 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
55
56
57 Conventions
58
59 All functions in this specification are mandatory.
60
61 <span class="h2"><a class="selflink" id="section-1" href="#section-1">1</a>. Introduction</span>
62
63 Pip is an internet protocol intended as the replacement for IP
64 version 4. Pip is a general purpose internet protocol, designed to
65 handle all forseeable internet protocol requirements. This
66 specification defines the Pip header processing for Routers and
67 Hosts.
68
69 The design of Pip is fundamentally different from that of previous
70 internetwork protocols. Pip is designed to be as general as
71 possible, but without significantly compromising performance.
72 Because of Pip's generality, it can handle forseeable routing and
73 addressing requirements. It is hoped that it will be able to handle
74 most if not all future routing and addressing requirements.
75
76 There are many detailed aspects of Pip that provide this generality
77 that are not discussed here. It is useful, however, to mention one
78 general aspect. That is, Pip strives to remove as much "functional
79 semantics" from the base specification as possible. Pip defines a
80 packet header and forwarding rules that can include many different
81 functional semantics (that is, routing, addressing, and flow
82 paradigms). Therefore, the reader may often find him or herself
83 asking "But how do you do foo with Pip?" The answer to this sort of
84 question belongs in companion documents to the basic Pip spec.
85
86 Pip can be thought of as a mechanism for triggering actions in hosts
87 and routers, just as a machine language can be thought of as a
88 mechanism for triggering actions in CPUs. The machine language has
89 no functional semantics outside of the specific actions it triggers
90 (move this register, write that memory location, etc.). But, the
91 machine language is a very powerful tool upon which functional
92 semantics are built. Likewise, Pip is a powerful tool upon which
93 routing, addressing, and flow functions can be built.
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108 <span class="grey">Francis [Page 2]</span></pre>
109 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-3" ></span>
110 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
111
112
113 <span class="h2"><a class="selflink" id="section-2" href="#section-2">2</a>. Pip Specification</span>
114
115 The Pip header is partitioned into three parts, the Initial Part, the
116 Transit Part, and the Options Part.
117
118
119 +===========================+
120 | Initial Part |
121 +===========================+
122 | Transit Part |
123 +===========================+
124 | Options Part |
125 +===========================+
126 | |
127 | Payload |
128 | |
129
130
131 Each part falls on a 32-bit boundary (as indicated by the double
132 lines shown), and the Transit Part falls on a 64 bit boundary.
133
134 The concept of tunneling in an integral part of Pip. Pip achieves
135 tunneling by encapsulating the Transit Part of the Pip header in
136 another Transit Part. Therefore, when tunneling, there is one
137 Transit Part for each (nested) tunnel:
138
139
140 +===========================+
141 | Initial Part |
142 +===========================+
143 | Transit Part |
144 +===========================+
145 | Transit Part |
146 +===========================+
147 .
148 .
149 .
150 +===========================+
151 | Transit Part |
152 +===========================+
153 | Options Part |
154 +===========================+
155
156
157 Because each Transit Part has only what is necessary for router
158 forwarding and handling, this method of tunneling is reasonably
159 efficient in terms of packet size.
160
161
162
163
164 <span class="grey">Francis [Page 3]</span></pre>
165 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-4" ></span>
166 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
167
168
169 <span class="h3"><a class="selflink" id="section-2.1" href="#section-2.1">2.1</a> Initial Part</span>
170
171 The Initial Part is formatted as shown in Figure 1.
172
173 length, in bits
174 +===========================+
175 | Version Number = 8 | 4
176 +---------------------------+
177 | Sub-Version | 4
178 +---------------------------+
179 | Options Offset | 8
180 +---------------------------+
181 | Options Contents | 8
182 +---------------------------+
183 | Options Present | 8
184 +===========================+
185 | Packet SubID | 16
186 +---------------------------+
187 | Protocol | 16
188 +===========================+
189 | Dest ID | 64
190 +===========================+
191 | Source ID | 64
192 +===========================+
193 | Payload Length | 32
194 +===========================+
195 | Host Version | 8
196 +---------------------------+
197 | Payload Offset | 8
198 +---------------------------+
199 | Hop Count | 16
200 +===========================+
201
202 Figure 1: Initial Part
203
204 An explanation of each field follows.
205
206 2.1.1 Version Numbers
207
208 The first octet is divided into two 4-bit fields, the Version and the
209 Sub-Version. The Version field is set to be 8, and is meant to be
210 version 8 of IP. (As of this writing, this is an experimental number
211 assigned for development of Pip.) Thus, all encapsulation schemes
212 defined for IP can work for Pip as well.
213
214 As long as the Version field is 8, the Initial Part and Options Part
215 of the Pip Header is as specified in this standard. (In other words,
216 the Sub-Version field refers only to the Transit Part.)
217
218
219
220 <span class="grey">Francis [Page 4]</span></pre>
221 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-5" ></span>
222 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
223
224
225 By doing this, we allow the Transit Part of the Pip Header to change
226 completely without necessarily requiring a host to understand the new
227 Transit Part. If a host receives a Pip header with a Version number
228 of 8 and an unknown Sub-version number, the host does not try to
229 parse the Transit Part at all, rather it processes only the Initial
230 Part and the Options Part. (By using the Pip Header Protocol to
231 format Pip Headers, a host can be made to formulate the right Transit
232 Part, even though the host doesn't understand the semantics of the
233 Transit Part. This allows radical migration of the Transit Part
234 while potentially not requiring changes to hosts.)
235
236 If a host or a router receives a packet with an unknown Version
237 number, the packet is silently discarded.
238
239 The Sub-Version field is set to the value 0 for the version of Pip
240 defined in this specification. As long as the Sub-Version number is
241 0, the Transit Part is as specified in this standard. Any packet
242 received by a router with a Version number of 8 and an unknown Sub-
243 Version number is silently discarded.
244
245 2.1.2 Options Offset
246
247 The Options Offset indicates the position of the Options Part. The
248 unit of measure of the Options Offset is 32-bit words, counting the
249 first word of the Pip Header as word 0.
250
251 2.1.3 Options Contents
252
253 This field indicates how the Options Present field should be
254 interpreted. Each bit of the Options field indicates if each of up
255 to eight options is present in the Options Part. The Options
256 Contents field indirectly indicates which option each bit of the
257 Options Present field refers to. We say indirectly because the
258 mapping referred to by the Options Contents field is stored locally.
259 In other words, without additional information (the mapping), it is
260 not possible to examine the Options Contents field and know what
261 option each bit of the Options Present field refers to.
262
263 Any of 256 possible Options Contents values can be active at a given
264 time. (Note that the means by which the meaning of the Options
265 Contents values are assigned and conveyed to routers and hosts is
266 outside the scope of this specification.)
267
268
269
270
271
272
273
274
275
276 <span class="grey">Francis [Page 5]</span></pre>
277 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-6" ></span>
278 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
279
280
281 2.1.4 Options Present
282
283 This field indicates which of the Options indicated by the Options
284 Contents field are actually present in the Options Part. Each bit of
285 this field refers to a single option type. The mapping of each bit
286 to its' option type is determined by the Options Contents field.
287
288 For instance, assume that the Options Contents field indicates that
289 bit 0 of the Options Present field refers to the PDN Address option,
290 that bit 1 of the Options Present field refers to the foo option, and
291 that bit 2 of the Options Present field refers to the Fragmentation
292 option. (As of this writing, there is only one option. Until there
293 are more than eight options, there is no need to define more than one
294 Options Contents values.)
295
296 In this case, a value of 101 in the Options Present field indicates
297 that the PDN Address and Fragmentation options are present in the
298 Options Part, and that the foo option is not present.
299
300 Note that an Options Present value of 0 indicates that there are no
301 options present, regardless of the value of the Options Contents
302 field. Note also that no more than 8 options, not including the
303 default first option (the Options Descriptor), can be present in any
304 Options Part.
305
306 The Options Contents/Options Present method of processing options
307 allows for efficient processing of options. First, a router can
308 ignore any options that may be present but that do not impact it (for
309 instance, a router not attached to a PDN need not consider the PDN
310 Address option). Second, the desired option can be very quickly
311 retrieved, because the first option, the Options Descriptor option,
312 contains the offset of each of the up to eight options indicated by
313 the Options Present field.
314
315 2.1.5 Packet SubID
316
317 This field is used by Pip hosts to correctly associate received PCMP
318 messages with local control blocks. This is necessary because the
319 semantics of the Transit Part can change while a packet is in
320 transit. Therefore, a router sending a PCMP message cannot
321 necessarily provide all of the information needed by the Pip host to
322 correctly identify the context of the received message (that is,
323 which "packet flow" it is identified with).
324
325 A PCMP message uses the Protocol, Source ID, Dest ID, and Packet
326 SubID to define the PCMP messages context. It is not sufficient to
327 use just Protocol, Source ID, and Dest ID, because two hosts running
328 the same protocol between them may have multiple "flows", for
329
330
331
332 <span class="grey">Francis [Page 6]</span></pre>
333 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-7" ></span>
334 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
335
336
337 instance, a data flow, a video flow, and an audio flow in the case of
338 multi-media. Each flow may have a different Transit Part, and take
339 different paths. Therefore, the Packet SubID field is needed to
340 further differentiate.
341
342 2.1.6 Protocol
343
344 Indicates the protocol header found in the payload. The values for
345 this field are the same as those used for IPv4.
346
347 2.1.7 Dest ID
348
349 The Dest ID field indicates the Pip ID of the final recipient of the
350 Pip packet. This field is examined by both hosts and routers.
351
352 When a Pip System processes the Routing Directive (RD), it may
353 determine that it needs to examine the Dest ID for further
354 processing. This may happen both when a host or router receives a
355 Pip packet destined for itself, or when a router receives a packet
356 that should be forwarded based on Dest ID (as indicated by the RD).
357
358 When a Pip system determines at forwarding time that a packet is
359 destined for itself, it checks the Dest ID to verify if that packet
360 is destined for it. If the complete Dest ID matches one of its own
361 Pip IDs, then the packet is for it, and is passed to the layer
362 indicated by the Protocol field (in the Host Part). (The Pip system
363 may of course wish to check a security option before passing a packet
364 to an upper layer.)
365
366 If the complete Dest ID field does not match one of its own IDs, then
367 an ID/RD Mismatch PCMP message is sent to the source of the packet,
368 as indicated by the Source ID and potentially source information in
369 the RD. The purpose of this message is to flush the ID to RD binding
370 in the source Pip host.
371
372 2.1.8 Source ID
373
374 This is the Pip ID of the source of the packet. It is passed to
375 upper layers for the purposes of identifying the context for the
376 packet.
377
378 2.1.9 Payload Length
379
380 The Payload Length gives the length of the Pip packet payload in
381 units of 8 bits. The Payload Length does not include the length of
382 the Pip header.
383
384
385
386
387
388 <span class="grey">Francis [Page 7]</span></pre>
389 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-8" ></span>
390 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
391
392
393 2.1.10 Host Version
394
395 The Host Version field indicates what "version" of Pip software the
396 sending host has implemented. This is to allow a host to inform a
397 router which ancillary protocols/messages the host is able to accept.
398 It is envisioned that over time, new host functions will be
399 developed. Different hosts will install these new functions at
400 different times. This field allows routers to know what functions
401 the host can and cannot handle.
402
403 2.1.11 Payload Offset
404
405 The Payload Offset indicates the position of the Payload Part. The
406 unit of measure of the Payload Offset is 32-bit words, counting the
407 first word of the Pip Header as word 0.
408
409 If a Pip system encapsulates a Transit Part in another Transit Part,
410 then the Payload Offset is increased by the length of the new Transit
411 Part.
412
413 2.1.12 Hop Count
414
415 The Hop Count is decremented by every router that forwards the Pip
416 packet. If a system receives a Pip header with a Hop Count equal to
417 0, and is not the recipient of the packet, then the packet is
418 discarded and a PCMP Destination Unreachable is routed to the system
419 indicated by the Routing Directive. (In other words, a host can
420 legally receive a Transit Part with a Hop Count of 0, and indeed a
421 host doesn't look at the Hop Count field upon reception.)
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444 <span class="grey">Francis [Page 8]</span></pre>
445 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-9" ></span>
446 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
447
448
449 <span class="h3"><a class="selflink" id="section-2.2" href="#section-2.2">2.2</a> Transit Part</span>
450
451 The Transit Part is formatted as shown in Figure 2.
452
453
454 length, in bits
455 +===========================+
456 | Reserved | 16
457 +---------------------------+
458 | Transit Part Offset | 8
459 +---------------------------+
460 | HD Contents | 8
461 +===========================+
462 | Handling Directive (HD) | 32
463 ---------------+===========================+
464 ^ | FTIF Offset | 8
465 | +---------------------------+
466 | | RC Contents | 8
467 | +---------------------------+
468 | | Routing Context (RC) | 16
469 Routing +===========================+
470 | FTIF 1 | 16
471 Directive +---------------------------+
472 | | FTIF 2 | 16
473 | +---------------------------+
474 | .
475 | .
476 | .
477 | +---------------------------+
478 | | FTIF N | 16
479 | +---------------------------+
480 v | Padding | Variable
481 ---------------+===========================+
482
483 Figure 2: Transit Part
484
485 An explanation of each field follows.
486
487 2.2.1 Transit Part Offset
488
489 This field gives the position of the first word of the next Transit
490 Part. The unit of measure of the Transit Part Offset is 32-bit
491 words, counting the first word of the current Transit Part as word 0.
492 If there is no next Transit Part, then this field is written as all
493 0's.
494
495
496
497
498
499
500 <span class="grey">Francis [Page 9]</span></pre>
501 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-10" ></span>
502 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
503
504
505 2.2.2 HD Contents
506
507 The HD Contents field indicates how the Handling Directive (HD) field
508 should be interpreted. The HD field is divided into multiple fields,
509 each representing a different handling function. Each individual
510 field in the HD is called an HD Unit (HDU). The Options Contents
511 field indirectly indicates which HDUs are in the HD field, and where
512 they are. We say indirectly because the mapping referred to by the
513 HD Contents field is stored locally. In other words, without
514 additional information (the mapping), it is not possible to examine
515 the HD Contents field and know what the HDU locations are.
516
517 Any of 256 possible HD Contents values can be active at a given time.
518 (Note that the means by which the meaning of the HD Contents values
519 are assigned and conveyed to routers and hosts is outside the scope
520 of this specification.)
521
522 2.2.3 Handling Directive (HD)
523
524 The HD is a general purpose field used for the purpose of triggering
525 special packet handling by a Pip system. The HD field does not
526 influence a Pip router's next hop choice for a Pip packet, nor does
527 it influence a Pip host's determination as to whether the Pip packet
528 is destined for it. Examples of special packet handling would be
529 "low priority queueing", or "high priority discard", etc. (Note that
530 the Transit Options also influence "handling", in the sense that
531 handling is essentially defined here to mean "anything that is not
532 routing. The HD field, though, is intended for the most common types
533 of handling--handling that is expected to be in a significant
534 percentage of packets.)
535
536 Both hosts and routers use the HD field. (Hosts may make use of the
537 HD field for packet handling for both incoming and outgoing packets.)
538
539 There is a complete distinction between the syntax and the semantics
540 of the HD field. (This can be contrasted with, for instance, IP,
541 which couples the semantics and syntax of the TOS bits. That is, the
542 IP specification itself determines, to a first degree, how the TOS
543 bits are interpreted.) Each Pip system can modify the semantic
544 meaning of the HD, for instance, by increasing or decreasing the
545 queueing priority of a packet. This is called packet tagging.
546
547 From an abstract modeling perspective, the HD is handled as follows:
548
549 1. Extract the semantic meaning(s) (the handling instructions
550 associated with the HDUs) from the HD field. Transmitting Pip
551 hosts determine the semantic meaning by some other means, such as
552 the upper layer protocol. If the receiving system decapsulates
553
554
555
556 <span class="grey">Francis [Page 10]</span></pre>
557 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-11" ></span>
558 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
559
560
561 multiple Pip headers, then the HD semantics are extracted from the
562 lowest Pip header for which it is not the target (see example on
563 tunneling below).
564
565 2. Handle the Pip packet according to those instructions. In some
566 cases, it is possible that the Pip system does not understand the
567 semantics of one or more HDUs of the HD field. For each HDU whose
568 semantics are not understood, however, the pip system at least
569 knows whether to 1) pass the HDU on untouched, 2) set it to all
570 0s, 3) set it to all 1s, 4) discard the packet silently, or 5)
571 discard the packet with a PCMP HDU Not Understood packet.
572
573 3. Modify the semantic meaning if necessary. Note also that if the
574 Pip packet is replicated for multicast, each packet has its HD
575 semantics modified individually. .LP .in 3 2.2.4 Tunneling .LP
576 Consider two Pip systems, X and Y, separated by one or
577 intermediate Pip systems. X wishes to tunnel a Transit Part to Y.
578 Y is therefore the target system of the tunnel. A Transit Part He
579 arrives at X. In order to forward the Transit Part to Y, X
580 encapsulates He in another Transit Part, Hy. Y is the target
581 system for Transit Part Hy. X sets the HD of He to what it would
582 have been if Y was directly connected to X (that is, there were no
583 intermediate Pip systems between X and Y). Further, it is
584 intended that Y will derive its HD semantics from the HD of
585 Transit Part He, not Transit Part Hy. .sp .KS
586
587 ----0-----o-----o-----o-----o-----0----
588 X I J K L Y
589
590 Now consider the operation of Pip system L (the previous hop system
591 to Y). When L forwards the packet to Y, it may either decapsulate
592 the packet (in the knowledge that Y is the target for Hy), or not
593 decapsulate the packet. Either way, L derives its HD semantics from
594 the HD of Transit Part Hy.
595
596 If L does not decapsulate the Transit Part, then it is as though I,
597 J, K, and L are a "subnetwork" (albeit a Pip subnetwork), and Y is
598 stripping the "subnetwork" header (Hy) off before processing the true
599 Transit Part (He). If L does decapsulate the Transit Part, then,
600 from Y's perspective, it is essentially as though Y were directly
601 connected to X.
602
603
604
605
606
607
608
609
610
611
612 <span class="grey">Francis [Page 11]</span></pre>
613 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-12" ></span>
614 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
615
616
617 2.2.5 Routing Directive (RD)
618
619 The RD consists of the Routing Context (RC), the RC Contents, the
620 FTIF Offset, and a series of zero or more FTIFs (Forwarding Table
621 Index Fields). This series of FTIFs is called the FTIF Chain. The
622 sole purpose of the RD is to determine how to forward the Pip
623 packet--the RD does not influence handling in any way.
624
625
626 Figure 3 illustrates the decision process for forwarding the Pip
627 packet.
628
629 +---------+(next level RC)
630 (decapsulate)| |
631 | v
632 |&lt;--------RC-----------------&gt;FIB
633 | / | | IF Offset)
634 | | |
635 | | v
636 |&lt;------|---FTIF-------------&gt;FIB
637 | | / :
638 | |&lt;- :(repeatedly...)
639 | | :
640 | | v
641 |&lt;------|---FTIF-------------&gt;FIB
642 | / |
643 |&lt;- |
644 v v
645 DestID--------------&gt;FIB
646
647 Figure 3: Forwarding Process
648
649
650 Figure 3 is interpreted as follows. The FIB is the Forwarding
651 Information Block. The FIB contains all the information needed to
652 forward a packet, and may contain multiple next hop (for multicast).
653 This information includes 1) the outgoing interface, 2) how to
654 encapsulate the packet, including lower-layer address(es) (the
655 lower-layer address(es) along with the outgoing interface determine
656 the next hop Pip system), 3) whether and how to tunnel, 4) how to
657 modify the semantics of the HD and RC, and how to modify the FTIF
658 Offset. The goal of the forwarding algorithm is to reach the
659 appropriate FIB.
660
661 The directed lines in Figure 3 start at the RC and, through various
662 possible paths, reach a FIB. These lines represent the various
663 information that can influence the forwarding decision (that is, the
664 FIB chosen). For instance, there is no way to reach a FIB without
665
666
667
668 <span class="grey">Francis [Page 12]</span></pre>
669 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-13" ></span>
670 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
671
672
673 first examining the information in the RC. However, it is possible
674 to identify a FIB by considering only the information in the RC (as
675 indicated by the directed line leading directly right from the RC).
676 Based on the information in the RC, it is also possible to determine
677 that the Transit Part must be decapsulated, and 1) the RC of the next
678 Transit Part be processed (the line leading directly left), 2) the
679 FTIF indicated by the FTIF Offset is processed (the line leading down
680 and right), or 3) the Dest ID is processed (the line leading down and
681 lest).
682
683 Likewise, when considering the value of an FTIF (in addition to all
684 information already considered), the resulting action may be that 1)
685 a FIB is identified, 2) the Transit Part is decapsulated, 3) the
686 subsequent FTIF is processed, or 4) the Dest ID is processed.
687
688 The RC is handled similarly to the HD. The RC Contents field
689 indicates how the RC should be interpreted. While the RC is
690 constructed similarly to the HD in the sense that it consists of
691 multiple fields, the RC can be interpreted as a flat field in-so-far
692 as forwarding a Pip packet is concerned, whereas the HD cannot.
693
694 Thus, in a mechanical sense, the RC Contents can be viewed as an
695 index into a table that returns a pointer to another table (an
696 rcTable), which is indexed by the RC itself. (Or, the combined RC
697 Contents/RC can be viewed as a single large index into a single
698 table, etc.)
699
700 The FTIF Offset field indicates which FTIF is active. The active
701 FTIF is the one that is used to index the forwarding table indicated
702 by the RC Contents/RC. An FTIF Offset value of 0 means that the
703 first FTIF is active, an FTIF Offset value of 1 means that the second
704 FTIF is active, and so on. If there are no FTIFs, then the FTIF
705 Offset has no meaning, and can be any value. In this case, the RC
706 field itself will indicate how to forward the packet.
707
708 The FTIF Chain is padded out to a 32-bit boundary. Note that there
709 can be more than 16 bits of padding (for instance, if it is desirable
710 to pad out to a 64-bit boundary). The padding is ignored upon
711 receipt, and can be transmitted as any value (that is, it does not
712 have to be any specific pattern of 0's or 1's).
713
714 Note that a single "number" in the FTIF chain may in fact be more
715 than 16 bits in length. In this case, the number can be encoded as
716 multiple FTIFs with no loss of generality. It is only required that
717 in all cases a multiple FTIF number be distinguishable from a single
718 FTIF number.
719
720
721
722
723
724 <span class="grey">Francis [Page 13]</span></pre>
725 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-14" ></span>
726 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
727
728
729 2.2.6 Router RD Forwarding Algorithm
730
731 This section describes the forwarding algorithm for a Pip router.
732
733 1. Using the value of the RC field as an index, retrieve one of the
734 following instructions (steps 2 - 5) from the rcTable determined
735 by the RC Contents.
736
737 2. If the instruction is decapsulate, then decapsulate the Transit
738 Part and re-execute step 1 using the next Transit Part.
739
740 3. If the instruction is forward, then retrieve the associated
741 Forwarding Information Block (FIB), and go to step 12.
742
743 4. If the instruction is to examine the Dest ID, then retrieve the
744 FIB associated with the Dest ID, and go to step 12.
745
746 5. If the instruction is to examine the FTIF Chain, then retrieve the
747 forwardingTable indicated by the rcTable entry, and continue on to
748 step 6.
749
750 6. Using the value of the currently active FTIF (this is the FTIF
751 indicated by the FTIF Offset if this is the first FTIF examined)
752 as an index, retrieve one or more of the following instructions
753 (steps 7 - 10) from the forwardingTable identified in step 5 or
754 step 10.
755
756 7. If the instruction is decapsulate, then decapsulate the Pip header
757 and re-execute step 1 using the new header (this is the same as
758 step 2).
759
760 8. If the instruction is forward, then (possibly additionally)
761 retrieve the associated FIB, and go to step 12 (this is the same
762 as step 3).
763
764 9. If the instruction is to examine the Dest ID, then retrieve the
765 FIB associated with the Dest ID and go to step 12 (this is the
766 same as step 4).
767
768 10. If the instruction is to examine the next FTIF, then, according
769 to the information in the current forwardingTable entry, modify
770 the current FTIF and choose a new forwardingTable.
771
772 11. Make the next FTIF the current FTIF and go to step 6.
773
774 12. The FIB contains a set of potential recipients for the Pip
775 packet, including next hop Pip systems (both directly connected
776 and at the end of Pip tunnels) and the upper layer of the local
777
778
779
780 <span class="grey">Francis [Page 14]</span></pre>
781 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-15" ></span>
782 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
783
784
785 system. Taking into consideration 1) the incoming interface, 2)
786 the previous hop Pip system if known (as determined by the
787 lower-layer source address and incoming interface), and 3)
788 potentially other local information (such as congestion on
789 outgoing queues), prune the set of potential recipients. (This
790 may result in no pruning having taken place or in every potential
791 next hop having been pruned.)
792
793 13. For each remaining next hop, format a Pip header by modifying a)
794 the RC, b) the current FTIF, c) the FTIF Offset (to point to 1)
795 the FTIF pointed to in the received RD, 2) the current FTIF, 3)
796 the Nth FTIF counting from the 0th FTIF, or 4) the Nth FTIF
797 counting forwards or backwards from the current FTIF) and d) any
798 Pip header encapsulations, according to the information in the
799 FIB, and transmit the packet to the recipient (either a next hop
800 or upper layer).
801
802 2.3 Options Part
803
804 The Option Part is formatted as shown in Figure 4.
805
806
807 +===========================+
808 | Options Descriptor | 64
809 +===========================+
810 | Option 2 | Variable
811 +===========================+
812 | Option 3 | Variable
813 +===========================+
814 .
815 .
816 .
817 +===========================+
818 | Option N | Variable
819 +===========================+
820
821
822 Figure 4: Options Part
823
824 Every Option is at least one 32-bit word in length, and ends on a
825 32-bit word boundary. Because the type of each option is known from
826 the Options Contents field, there is no need to indicate the option
827 type in the options field themselves. Thus, there is no common
828 format among the options--each option has its own format. The
829 individual options are defined in another specification.
830
831
832
833
834
835
836 <span class="grey">Francis [Page 15]</span></pre>
837 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-16" ></span>
838 <span class="grey"><a href="./rfc1622">RFC 1622</a> Pip Header Processing May 1994</span>
839
840
841 2.3.1 Options Descriptor
842
843 The Options Descriptor option gives the offset of each option in the
844 Options Part. The Options Descriptor consists of eight eight-bit
845 Option Position fields, each of which gives the position of up to
846 eight options (there can be no more than 8 Options Part). Each of
847 the Option Position fields correspond to one of the bits in the
848 Options Present field. The unit of measure of each Option Position
849 is 32-bit words, counting the first word of the Options Part as word
850 0. The high order Option Position field corresponds to the high
851 order bit in the Options Present field.
852
853 Security Considerations
854
855 Security issues are not discussed in this memo.
856
857 Author's Address
858
859 Paul Francis
860 NTT Software Lab
861 3-9-11 Midori-cho Musashino-shi
862 Tokyo 180 Japan
863
864 Phone: +81-422-59-3843
865 Fax +81-422-59-3765
866 EMail: [email protected]
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892 Francis [Page 16]
893 </pre>