ARMWARE RFC Archive <- RFC Index (4801..4900)

RFC 4815

Updates RFC 3095, RFC 3241, RFC 3843, RFC 4019, RFC 4362

Network Working Group                                       L-E. Jonsson
Request for Comments: 4815                                   K. Sandlund
Updates: 3095, 3241, 3843, 4019, 4362                       G. Pelletier
Category: Standards Track                                      P. Kremer
                                                           February 2007

                   RObust Header Compression (ROHC):
               Corrections and Clarifications to RFC 3095

Status of This Memo

   This document specifies an Internet standards track protocol for the
   Internet community, and requests discussion and suggestions for
   improvements.  Please refer to the current edition of the "Internet
   Official Protocol Standards" (STD 1) for the standardization state
   and status of this protocol.  Distribution of this memo is unlimited.

Copyright Notice

   Copyright (C) The IETF Trust (2007).

Abstract

   RFC 3095 defines the RObust Header Compression (ROHC) framework and
   profiles for IP (Internet Protocol), UDP (User Datagram Protocol),
   RTP (Real-Time Transport Protocol), and ESP (Encapsulating Security
   Payload).  Some parts of the specification are unclear or contain
   errors that may lead to misinterpretations that may impair
   interoperability between different implementations.  This document
   provides corrections, additions, and clarifications to RFC 3095; this
   document thus updates RFC 3095.  In addition, other clarifications
   related to RFC 3241 (ROHC over PPP), RFC 3843 (ROHC IP profile) and
   RFC 4109 (ROHC UDP-Lite profiles) are also provided.

Jonsson, et al.             Standards Track                     [Page 1]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

Table of Contents

   1. Introduction and Terminology ....................................3
   2. CRC Calculation and Coverage ....................................4
      2.1. CRC Calculation ............................................4
      2.2. Padding Octet and CRC Calculations .........................4
      2.3. CRC Coverage in CRC Feedback Options .......................5
      2.4. CRC Coverage of the ESP NULL Header ........................5
   3. Mode Transition .................................................5
      3.1. Feedback During Mode Transition to U- and O-Mode ...........5
           3.1.1. Mode Transition Procedures Allowing Sparse Feedback .6
           3.1.2. Transition from Reliable to Optimistic Mode .........7
           3.1.3. Transition to Unidirectional Mode ...................8
      3.2. Feedback During Mode Transition ............................8
      3.3. Packet Decoding During Mode Transition .....................9
   4. Timestamp Encoding ..............................................9
      4.1. Encoding Used for Compressed TS Bits .......................9
      4.2. (De)compression of TS without Transmitted TS Bits .........10
      4.3. Interpretation Intervals for TS Encoding ..................11
      4.4. Scaled RTP Timestamp Encoding .............................11
           4.4.1. TS_STRIDE for Scaled Timestamp Encoding ............11
           4.4.2. TS Wraparound with Scaled Timestamp Encoding .......12
           4.4.3. Algorithm for Scaled Timestamp Encoding ............12
      4.5. Recalculating TS_OFFSET ...................................14
      4.6. TS_STRIDE and the Tsc Flag in Extension 3 .................14
      4.7. Using Timer-Based Compression .............................15
   5. List Compression ...............................................15
      5.1. CSRC List Items in RTP Dynamic Chain ......................15
      5.2. Multiple Occurrences of the CC Field ......................15
      5.3. Bit Masks in List Compression .............................16
      5.4. Headers Compressed with List Compression ..................16
      5.5. ESP NULL Header List Compression ..........................17
      5.6. Translation Tables and Indexes for IP Extension Headers ...17
      5.7. Reference List ............................................17
      5.8. Compression of AH and GRE Sequence Numbers ................18
   6. Updating Properties ............................................19
      6.1. Implicit Updates ..........................................19
      6.2. Updating Properties of UO-1* ..............................20
      6.3. Context Updating Properties for IR Packets ................20
      6.4. RTP Padding Field (R-P) in Extension 3 ....................20
      6.5. RTP eXtension bit (X) in dynamic part .....................21
   7. Context management and CID/context Reuse .......................21
      7.1. Persistence of Decompressor Contexts ......................21
      7.2. CID/Context Reuse .........................................21
           7.2.1. Reusing a CID/Context with the Same Profile ........22
           7.2.2. Reusing a CID/Context with a Different Profile .....23
   8. Other Protocol Clarifications ..................................23
      8.1. Meaning of NBO ............................................23

Jonsson, et al.             Standards Track                     [Page 2]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

      8.2. IP-ID .....................................................23
      8.3. Extension-3 in UOR-2* Packets .............................24
      8.4. Multiple Occurrences of the M Bit .........................24
      8.5. Multiple SN options in one feedback packet ................24
      8.6. Multiple CRC Options in One Feedback Packet ...............25
      8.7. Responding to Lost Feedback Links .........................25
      8.8. UOR-2 in Profile 0x0002 (UDP) and Profile 0x0003 (ESP) ....25
      8.9. Sequence Number LSB's in IP Extension Headers .............25
      8.10. Expecting UOR-2 ACKs in O-Mode ...........................26
      8.11. Context Repairs, TS_STRIDE and TIME_STRIDE ...............26
   9. ROHC Negotiation ...............................................27
   10. PROFILES Sub-option in ROHC-over-PPP ..........................27
   11. Constant IP-ID Encoding in IP-only and UPD-Lite Profiles ......27
   12. Security Considerations .......................................28
   13. Acknowledgments ...............................................28
   14. References ....................................................28
      14.1. Normative References .....................................28
      14.2. Informative References ...................................29
   Appendix A. Sample CRC Algorithm ..................................30

1.  Introduction and Terminology

   RFC 3095 [1] defines the RObust Header Compression (ROHC) framework
   and profiles for IP (Internet Protocol) [8][9], UDP (User Datagram
   Protocol) [10], RTP (Real-Time Transport Protocol) [11], and ESP
   (Encapsulating Security Payload) [12].  During implementation and
   interoperability testing of RFC 3095, some ambiguities and common
   misinterpretations have been identified, as well as a few errors.

   This document summarizes identified issues and provides corrections
   needed for implementations of RFC 3095 to interoperate, i.e., it
   constitutes an update to RFC 3095.  This document also provides other
   clarifications related to common misinterpretations of the
   specification.  References to RFC 3095 should, therefore, also
   include this document.

   In addition, some clarifications and corrections are also provided
   for RFC 3241 (ROHC over PPP) [2], RFC 3843 (ROHC IP-only profile)
   [4], and RFC 4019 (ROHC UDP-Lite profiles) [5], which are thus also
   updated by this document.  Furthermore, RFC 4362 (ROHC Link-Layer
   Assisted Profile) [7] is implicitly updated by this document, since
   RFC 4362 is also based on RFC 3095.

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in RFC 2119 [6].

Jonsson, et al.             Standards Track                     [Page 3]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   When a section of this document makes formal corrections, additions
   or invalidations to text in RFC 3095, this is clearly summarized.
   The text from RFC 3095 that is being addressed is given and labeled
   "INCOMPLETE", "INCORRECT", or "INCORRECT AND INVALIDATED", followed
   by the correct text labeled "CORRECTED", where applicable.  When text
   is added that does not simply correct text in previous
   specifications, it is given with the label "FORMAL ADDITION".

   In this document, a reference to a section in RFC 3095 [1] is written
   as RFC 3095-Section <number>.

2.  CRC Calculation and Coverage

2.1.  CRC Calculation

   RFC 3095-Section 5.9 defines polynomials for 3-, 7-, and 8-bit Cyclic
   Redundancy Checks (CRCs), but it does not specify what algorithm is
   used.  The 3-, 7- and 8-bit CRCs are calculated using the CRC
   algorithm defined in [3].

   A Perl implementation of the algorithm can be found in Appendix A of
   this document.

2.2.  Padding Octet and CRC Calculations

   RFC 3095-Section 5.9.1 is incomplete, as it does not mention how to
   handle the padding octet in CRC calculations for IR and IR-DYN
   packets.  Padding isn't meant to be a meaningful part of a packet and
   is not included in the CRC calculation.  As a result, the CRC does
   not cover the Add-CID octet for CID 0, either.

   INCOMPLETE RFC 3095 TEXT (RFC 3095-Section 5.9.1):

      "The CRC in the IR and IR-DYN packet is calculated over the entire
       IR or IR-DYN packet, excluding Payload and including CID or any
       Add-CID octet."

   CORRECTED TEXT:

      "The CRC in the IR and IR-DYN packet is calculated over the entire
       IR or IR-DYN packet, excluding Payload, Padding and including CID
       or any Add-CID octet, except for the add-CID octet for CID 0."

Jonsson, et al.             Standards Track                     [Page 4]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

2.3.  CRC Coverage in CRC Feedback Options

   RFC 3095-Section 5.7.6.3 is incomplete, as it does not mention how
   the "size" field is handled when calculating the 8-bit CRC used in
   the CRC feedback option.  Since the "size" field is an extension of
   the "code" field, it must be treated in the same way.

   INCOMPLETE RFC 3095 TEXT (RFC 3095-Section 5.7.6.3):

      "The CRC option contains an 8-bit CRC computed over the entire
       feedback payload, without the packet type and code octet, but
       including any CID fields, using the polynomial of section 5.9.1."

   CORRECTED TEXT:

      "The CRC option contains an 8-bit CRC computed over the entire
       feedback payload including any CID fields but excluding the
       packet type, the 'Size' field and the 'Code' octet, using the
       polynomial of Section 5.9.1."

2.4.  CRC Coverage of the ESP NULL Header

   RFC 3095-Section 5.8.7 gives the CRC coverage of the ESP NULL [13]
   header as "Entire ESP header".  This must be interpreted as including
   only the initial part of the header (i.e., Security Parameter Index
   (SPI) and sequence number), and not the trailer part at the end of
   the payload.  Therefore, the ESP NULL header has the same CRC
   coverage as the ESP header used in the ESP profile (RFC 3095-Section
   5.7.7.7).

3.  Mode Transition

3.1.  Feedback During Mode Transition to U- and O-Mode

   RFC 3095-Section 5.6.1 states that during mode transitions, while the
   D_TRANS parameter is I, the decompressor sends feedback for each
   received packet.  This restrictive behavior prevents the decompressor
   from using a sparse feedback algorithm during mode transitions.

   To reduce transmission overhead and computational complexity
   (including CRC calculation) associated with feedback packets sent for
   each decompressed packet during mode transition, a decompressor MAY
   be implemented with slightly modified mode transition procedures
   compared to those defined in [1], as described in this section.

   These enhanced procedures should be considered only as a possible
   improvement to a decompressor implementation, since interoperability
   is not affected in any way.  A decompressor implemented according to

Jonsson, et al.             Standards Track                     [Page 5]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   the optimized procedures will interoperate with an RFC 3095
   compressor, as well as a decompressor implemented according to the
   procedures described in RFC 3095.

3.1.1.  Mode Transition Procedures Allowing Sparse Feedback

   The purpose of these enhanced transition procedures is to allow the
   decompressor to sparsely send feedback for packets decompressed
   during the second half of the transition procedure, i.e., after an
   appropriate IR/IR-DYN/UOR-2 packet has been received from the
   compressor.  This is achieved by allowing the decompressor transition
   parameter (D_TRANS) to be set to P (Pending) at that stage, as shown
   in the transition diagrams of Sections 3.1.2 and 3.1.3 below.

   This enhanced transition, where feedback need not be sent for every
   decompressed packet, does however introduce some considerations in
   case feedback messages would be lost.  Specifically, there is a risk
   for a deadlock situation when a transition from R-mode is performed;
   if no feedback message successfully reaches the compressor, the
   transition is never completed.  For transition between U-mode and
   O-mode, there is also a small risk for reduced compression
   efficiency.

   To avoid this, the decompressor MUST continue to send feedback at
   least periodically, as well as when in a Pending transition state.
   This is equivalent to enhancing the definition of the D_TRANS
   parameter in RFC 3095-Section 5.6.1, to include the definition of a
   Pending state:

   -  D_TRANS:
      Possible values for the D_TRANS parameter are (I)NITIATED,
      (P)ENDING, and (D)ONE.  D_TRANS MUST be initialized to D, and a
      mode transition can be initiated only when D_TRANS is D.  While
      D_TRANS is I, the decompressor sends a NACK or ACK carrying a CRC
      option for each packet received.  When D_TRANS is set to P, the
      decompressor does not have to send a NACK or ACK for each packet
      received, but it MUST continue to send feedback with some
      periodicity, and all feedback packets sent MUST include the CRC
      option.  This ensures that all mode transitions will be completed
      also in case of feedback losses.

   The modifications affect transitions to Optimistic and Unidirectional
   modes of operation (i.e., the transitions described in RFC 3095-
   Section 5.6.5 and RFC 3095-Section 5.6.6) and make those transition
   diagrams more consistent with the diagram describing the transition
   to R-mode.

Jonsson, et al.             Standards Track                     [Page 6]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

3.1.2.  Transition from Reliable to Optimistic Mode

   The enhanced procedure for transition from Reliable to Optimistic
   mode is shown below:

             Compressor                     Decompressor
            ----------------------------------------------
                  |                               |
                  |        ACK(O)/NACK(O) +-<-<-<-| D_TRANS = I
                  |       +-<-<-<-<-<-<-<-+       |
      C_TRANS = P |-<-<-<-+                       |
      C_MODE = O  |                               |
                  |->->->-+ IR/IR-DYN/UOR-2(SN,O) |
                  |       +->->->->->->->-+       |
                  |->-..                  +->->->-| D_TRANS = P
                  |->-..                          | D_MODE = O
                  |           ACK(SN,O)   +-<-<-<-|
                  |       +-<-<-<-<-<-<-<-+       |
      C_TRANS = D |-<-<-<-+                       |
                  |                               |
                  |->->->-+  UO-0, UO-1*          |
                  |       +->->->->->->->-+       |
                  |                       +->->->-| D_TRANS = D
                  |                               |

Jonsson, et al.             Standards Track                     [Page 7]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

3.1.3.  Transition to Unidirectional Mode

   The enhanced procedure for transition to Unidirectional mode is shown
   on the following figure:

                 Compressor                     Decompressor
                ----------------------------------------------
                  |                               |
                  |        ACK(U)/NACK(U) +-<-<-<-| D_TRANS = I
                  |       +-<-<-<-<-<-<-<-+       |
      C_TRANS = P |-<-<-<-+                       |
      C_MODE = U  |                               |
                  |->->->-+ IR/IR-DYN/UOR-2(SN,U) |
                  |       +->->->->->->->-+       |
                  |->-..                  +->->->-| D_TRANS = P
                  |->-..                          |
                  |           ACK(SN,U)   +-<-<-<-|
                  |       +-<-<-<-<-<-<-<-+       |
      C_TRANS = D |-<-<-<-+                       |
                  |                               |
                  |->->->-+  UO-0, UO-1*          |
                  |       +->->->->->->->-+       |
                  |                       +->->->-| D_TRANS = D
                  |                               | D_MODE= U

3.2.  Feedback During Mode Transition

   RFC 3095-Section 5.6.1 states that feedback is always used during
   mode transitions.  However, the text then continues by making
   concrete applications of the rule in an inconsistent way, making it
   unclear when CRCs are used.  Further, the text does not define how
   the compressor should act during mode transitions based on feedback
   not protected by CRCs, i.e., whether or not to carry out mode
   transition actions.  The proper behavior from the compressor is to
   perform any action related to mode transitions only when the feedback
   is protected by the CRC option.

   INCOMPLETE RFC 3095 TEXT (RFC 3095-Section 5.6.1):

      "As a safeguard against residual errors, all feedback sent during
       a mode transition MUST be protected by a CRC, i.e., the CRC
       option MUST be used."

   CORRECTED TEXT:

       "As a safeguard against residual errors, all feedback sent by the
       decompressor during a mode transition MUST be protected by a CRC,
       i.e., the CRC option MUST be used.  The compressor MUST ignore

Jonsson, et al.             Standards Track                     [Page 8]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

       feedback information related to mode transition if the feedback
       is not protected by the CRC option."

   One more related issue that requires clarifications comes from the
   following text at the end of RFC 3095-Section 5.6.1:

      "While D_TRANS is I, the decompressor sends a NACK or ACK carrying
       a CRC option for each received packet."

   However, RFC 3095-Section 5.5.2.2 already stated that for R-mode,
   feedback is never sent for packets that do not update the context,
   i.e., for packets that do not carry a CRC, such as R-0 and R-1*.

   This means that when D_TRANS=I during mode transition, a decompressor
   operating in R-mode sends an acknowledgement for each packet it
   receives and MUST use the sequence number that corresponds to the
   packet that last updated the context, i.e., the decompressor MUST NOT
   use the sequence number of the R-0 or the R-1* packet.

3.3.  Packet Decoding During Mode Transition

   The purpose of a mode transition is to ensure that the compressor and
   the decompressor coherently move from one mode of operation to
   another using a three-way handshake.  At one point during the mode
   transition, the decompressor acknowledges the reception of one (or
   more) IR, IR-DYN or UOR-2 packet(s) that have mode bits set to the
   new mode.  Packets of type 0 or type 1 that are received up to this
   point are decompressed using the old mode, while afterwards they are
   decompressed using the new mode.  If the enhanced transition
   procedures described in Section 3.1 are used, the setting of the
   D_TRANS parameter to P represents this breakpoint.  The successful
   decompression of a packet of type 0 or type 1 completes the mode
   transition.

4.  Timestamp Encoding

4.1.  Encoding Used for Compressed TS Bits

   RTP Timestamp (TS) values are always encoded using W-LSB encoding,
   both when sent scaled and unscaled.  When no TS bits are transmitted
   in a compressed packet, TS is always scaled.  If a compressed packet
   carries an Extension 3 and field(Tsc)=0, the compressed packet must
   thus always carry unscaled TS bits.  For TS values sent in Extension
   3, W-LSB encoded values are sent using the self-describing variable-
   length format (RFC 3095-Section 4.5.6), and this applies to both
   scaled and unscaled values.

Jonsson, et al.             Standards Track                     [Page 9]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

4.2.  (De)compression of TS without Transmitted TS Bits

   When ROHC RTP operates using its most efficient packet types, apart
   from packet type identification and the error detection CRC, only RTP
   sequence number (SN) bits are transmitted in RTP compressed headers.
   All other fields are then omitted either because they are unchanged
   or because they can be reconstructed through a function from the SN
   (i.e., by combining the transmitted SN bits with state information
   from the context).  Fields that can be inferred from the SN are the
   IP Identification (IP-ID) and the RTP Timestamp (TS).

   IP-ID compression and decompression, both with and without
   transmitted IP-ID bits in the compressed header, are well defined in
   RFC 3095-Section 4.5.5 (see Section 8.2).  For the TS field, however,
   RFC 3095 only defines how to decompress based on actual TS bits in
   the compressed header, either scaled or unscaled, but not how to
   infer the TS from the SN when there are no TS bits present in the
   compressed header.

   When no TS bits are received in the compressed header, the scaled TS
   value is reconstructed assuming a linear extrapolation from the SN,
   i.e., delta_TS = delta_SN * default-slope, where delta_SN and
   delta_TS are both signed integers.  RFC 3095-Section 5.7 defines the
   potential values for default-slope.

   INCOMPLETE RFC 3095 TEXT (RFC 3095-Section 5.7):

      "If value(Tsc) = 1, Scaled RTP Timestamp encoding is used before
       compression (see section 4.5.3), and default-slope(TS) = 1.

       If value(Tsc) = 0, the Timestamp value is compressed as-is, and
       default-slope(TS) = value(TS_STRIDE)."

   CORRECTED TEXT:

      "When a compressed header with no TS bits is received, the scaled
       TS value is reconstructed assuming a linear extrapolation from
       the SN, i.e., delta_TS = delta_SN * default-slope(TS).

       If value(Tsc) = 1, Scaled RTP Timestamp encoding is used before
       compression (see Section 4.5.3), and default-slope(TS) = 1.

       If value(Tsc) = 0, the Timestamp value is compressed as-is, and
       default-slope(TS) = value(TS_STRIDE).  If a packet with no TS
       bits is received with Tsc = 0, the decompressor MUST discard the
       packet."

Jonsson, et al.             Standards Track                    [Page 10]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   INCORRECT AND INVALIDATED RFC 3095 TEXT (Section RFC 3095-5.5.1.2):

       "For example, in a typical case where the string pattern has the
       form of non-SN-field = SN * slope + offset, one ACK is enough if
       the slope has been previously established by the decompressor
       (i.e., only the new offset needs to be synchronized).  Otherwise,
       two ACKs are required since the decompressor needs two headers to
       learn both the new slope and the new offset."

   Consequently, there is no other slope value than the default-slope,
   as defined in RFC 3095-Section 5.7.

4.3.  Interpretation Intervals for TS Encoding

   RFC 3095-Section 4.5.4 defines the interpretation interval, p, for
   timer-based compression of the RTP timestamp.  However, RFC 3095-
   Section 5.7 defines a different interpretation interval, which is
   defined as the interpretation interval to use for all TS values.  It
   is thus unclear which p-value to use, at least for timer-based
   compression.

   The way this should be interpreted is that the p-value differs
   depending on whether or not timer-based compression is enabled.

   For timer-based compression (TIME_STRIDE set to a non-zero value),
   the interpretation interval is:

      p = 2^(k-1) - 1 (as per RFC 3095-Section 4.5.4)

   Otherwise, the interpretation interval is:

      p = 2^(k-2) - 1 (as per RFC 3095-Section 5.7)

4.4.  Scaled RTP Timestamp Encoding

   This section redefines the algorithm for scaled RTP timestamp
   encoding, defined as a 5-step procedure in RFC 3095-Section 4.5.3.
   Two formal errors have been corrected, as described in sub-sections
   4.4.1 and 4.4.2 below, and the whole algorithm has been reworked to
   be more concise and to use well-defined terminology.  The resulting
   text can be found in 4.4.3 below.

4.4.1.  TS_STRIDE for Scaled Timestamp Encoding

   RFC 3095 defines the timestamp stride (TS_STRIDE) as the expected
   increase in the timestamp value between two RTP packets with
   consecutive sequence numbers.  TS_STRIDE is set by the compressor and

Jonsson, et al.             Standards Track                    [Page 11]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   explicitly communicated to the decompressor, and it is used as the
   scaling factor for scaled TS encoding.

   The relation between TS and TS_SCALED, given by the following
   equality in RFC 3095-Section 4.5.3, defines the mathematical meaning
   of TS_STRIDE:

      TS = TS_SCALED * TS_STRIDE + TS_OFFSET

   TS_SCALED is incompletely written as TS / TS_STRIDE in the
   compression step following the above core equality.  This formula is
   incorrect both because it excludes TS_OFFSET and because it would
   prevent a TS_STRIDE value of 0, which is an alternative not excluded
   by the definition or by the core equality above.  If "/" were a
   generally unambiguously defined operation meaning "the integral part
   of the result from dividing X by Y", the absence of TS_OFFSET could
   be explained, but the formula would still lack a proper output for
   TS_STRIDE equal to 0.  The formula of "2. Compression" is thus valid
   only with the following requirements:

     a) "/" means "the integral part of the result from dividing X by Y"

     b) TS_STRIDE>0 (TS is never sent scaled when TS_STRIDE=0)

4.4.2.  TS Wraparound with Scaled Timestamp Encoding

   RFC 3095-Section 4.5.3 states in points 4 and 5 that the compressor
   is not required to initialize TS_OFFSET at wraparound, but that it is
   required to increase the number of bits sent for the scaled TS value
   when there is a TS wraparound.  The decompressor is also required to
   detect and cope with TS wraparound, including updating TS_OFFSET.

   This method is not interoperable and not robust.  The gain is also
   insignificant, as TS wraparound happens very seldomly.  Therefore,
   the compressor should reinitialize TS_OFFSET upon TS wraparound, by
   sending an unscaled TS.

4.4.3.  Algorithm for Scaled Timestamp Encoding

   INCORRECT RFC 3095 TEXT (RFC 3095-Section 4.5.3):

     "1. Initialization: The compressor sends to the decompressor the
         value of TS_STRIDE and the absolute value of one or several TS
         fields.  The latter are used by the decompressor to initialize
         TS_OFFSET to (absolute value) modulo TS_STRIDE.  Note that
         TS_OFFSET is the same regardless of which absolute value is
         used, as long as the unscaled TS value does not wrap around;
         see 4) below.

Jonsson, et al.             Standards Track                    [Page 12]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

      2. Compression: After initialization, the compressor no longer
         compresses the original TS values.  Instead, it compresses the
         downscaled values: TS_SCALED = TS / TS_STRIDE.  The compression
         method could be either W-LSB encoding or the timer-based
         encoding described in the next section.

      3. Decompression: When receiving the compressed value of
         TS_SCALED, the decompressor first derives the value of the
         original TS_SCALED.  The original RTP TS is then calculated as
         TS = TS_SCALED * TS_STRIDE + TS_OFFSET.

      4. Offset at wraparound: Wraparound of the unscaled 32-bit TS will
         invalidate the current value of TS_OFFSET used in the equation
         above.  For example, let us assume TS_STRIDE = 160 = 0xA0 and
         the current TS = 0xFFFFFFF0.  TS_OFFSET is then 0x50 = 80.
         Then if the next RTP TS = 0x00000130 (i.e., the increment is
         160 * 2 = 320), the new TS_OFFSET should be 0x00000130 modulo
         0xA0 = 0x90 = 144.  The compressor is not required to re-
         initialize TS_OFFSET at wraparound.  Instead, the decompressor
         MUST detect wraparound of the unscaled TS (which is trivial)
         and update TS_OFFSET to TS_OFFSET = (Wrapped around unscaled
         TS) modulo TS_STRIDE"

   CORRECTED TEXT:

     "1. Initialization and updating of RTP TS scaling function:  The
         compressor sends to the decompressor the value of TS_STRIDE
         along with an unscaled TS.  These are both needed by the
         decompressor to initialize TS_OFFSET as hdr(TS) modulo
         field(TS_STRIDE).  Note that TS_OFFSET is the same for any TS
         as long as TS_STRIDE does not change and as long as the
         unscaled TS value does not wrap around; see 4) below.

      2. Compression: After initialization, the compressor no longer
         compresses the unscaled TS values.  Instead, it compresses the
         scaled values.  The compression method can be either W-LSB
         encoding or timer-based encoding.

      3. Decompression: When receiving a (compressed) TS_SCALED, the
         field is first decompressed, and the unscaled RTP TS is then
         calculated as TS = TS_SCALED * TS_STRIDE + TS_OFFSET.

      4. Offset at wraparound: If the value of TS_STRIDE is not equal to
         a power of two, wraparound of the unscaled 32-bit TS will
         change the value of TS_OFFSET.  When this happens, the
         compressor SHOULD reinitialize TS_OFFSET by sending unscaled
         TS, as in 1 above."

Jonsson, et al.             Standards Track                    [Page 13]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   INCORRECT AND INVALIDATED RFC 3095 TEXT (RFC 3095-Section 4.5.3):

      The entire point 5, i.e. the entire text starting from "5.
      Interpretation interval at wraparound ...", down to and including
      the block of text that starts with "Let a be the number of LSBs"
      and that ends with "...interpretation interval is b." is incorrect
      and is thus invalid.

4.5.  Recalculating TS_OFFSET

   TS can be sent unscaled if the TS value change does not match the
   established TS_STRIDE, but the TS_STRIDE might still stay unchanged.
   To ensure correct decompression of subsequent packets, the
   decompressor MUST therefore always recalculate TS_OFFSET (RTP TS
   modulo TS_STRIDE) when a packet with an unscaled TS value is
   received.

4.6.  TS_STRIDE and the Tsc Flag in Extension 3

   The Tsc flag in Extension 3 indicates whether or not TS is scaled.
   The value of the Tsc flag thus applies to all TS bits, as well as if
   there are no TS bits in the extension itself.  When TS is scaled, it
   is always scaled using context(TS_STRIDE).  The legend for Extension
   3 in RFC 3095-Section 5.7.5 incorrectly states that value(TS_STRIDE)
   is used for scaled TS.

   If TS_STRIDE is present in Extension 3, as indicated by the Tss flag
   being set, the compressed header SHOULD carry unscaled TS bits; i.e.,
   the Tsc flag SHOULD NOT be set when Tss is set since an unscaled TS
   is needed together with TS_STRIDE to recalculate the TS_OFFSET.  If
   TS_STRIDE is included in a compressed header with scaled TS, the
   decompressor must ignore and discard field(TS_STRIDE).

   INCORRECT RFC 3095 TEXT (RFC 3095-Section 5.7.5):

      "Tsc: Tsc = 0 indicates that TS is not scaled;
            Tsc = 1 indicates that TS is scaled according to section
             4.5.3, using value(TS_STRIDE).
             Context(Tsc) is always 1.  If scaling is not desired, the
             compressor will establish TS_STRIDE = 1."

   CORRECTED TEXT:

      "Tsc: Tsc = 0 indicates that TS is not scaled;
            Tsc = 1 indicates that TS is scaled according to Section
            4.5.3, using context(TS_STRIDE).

Jonsson, et al.             Standards Track                    [Page 14]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

            Context(Tsc) is always 1.  If scaling is not desired, the
            compressor will establish TS_STRIDE = 1.

            If field(Tsc) = 1, and if TSS = 1 (meaning that TS_STRIDE is
            present in the extension), field(TS_STRIDE) MUST be ignored
            and discarded."

   When the compressor re-establishes a new value for TS_STRIDE using
   Extension 3, it should send unscaled TS bits together with TS_STRIDE.

4.7.  Using Timer-Based Compression

   Timer-based compression of the RTP timestamp, as described in RFC
   3095-Section 4.5.4, may be used to reduce the number of transmitted
   timestamp bits (bytes) needed when the timestamp cannot be inferred
   from the SN.  Timer-based compression is only used for decompression
   of compressed headers that contains a TS field; otherwise, when no
   timestamp bits are present, the timestamp is linearly inferred from
   the SN (see Section 4.2 of this document).

   Whether or not to use timer-based compression is controlled by the
   TIME_STRIDE control field, which can be set by either an IR, an IR-
   DYN, or a compressed packet with Extension 3.  Before timer-based
   compression can be used, the decompressor has to inform the
   compressor (on a per-channel basis) about its clock resolution by
   sending a CLOCK feedback option for any CID on the channel.  The
   compressor can then initiate timer-based compression by sending (on a
   per-context basis) a non-zero TIME_STRIDE to the decompressor.  When
   the compressor is confident that the decompressor has received the
   TIME_STRIDE value, it can switch to timer-based compression.

5.  List Compression

5.1.  CSRC List Items in RTP Dynamic Chain

   RFC 3095-Section 5.7.7.6 defines the static and dynamic parts of the
   RTP header.  This section indicates a 'Generic CSRC list' field in
   the dynamic chain, which has a variable length (see RFC 3095-Section
   5.8.6).  This field is always at least one octet in size, even if the
   list is empty (as opposed to the CSRC list in the uncompressed RTP
   header, which is not present when the RTP CC field is set to 0).

5.2.  Multiple Occurrences of the CC Field

   The static and the dynamic parts of the RTP header are defined in RFC
   3095-Section 5.7.7.6.  In the dynamic part, a CC field indicates the
   number of CSRC items present in the 'Generic CSRC list'.  Another CC
   field also appears within the 'Generic CSRC list' (RFC 3095-Section

Jonsson, et al.             Standards Track                    [Page 15]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   5.8.6.1), because Encoding Type 0 is always used in the dynamic
   chain.  Both CC fields have the same meaning: the value of the CC
   field determines the number of XI items in the CSRC list for Encoding
   Type 0, and it is not used otherwise.  Therefore, the following
   applies:

   FORMAL ADDITION TO RFC 3095:

      "The first octet in the dynamic part of the RTP header contains a
       CC field, as defined in Section 5.7.7.6.  A second occurrence
       appears in the 'Generic CSRC list', which is also in the dynamic
       part of the RTP header, where Encoding Type 0 is used according
       to the format defined in RFC 3095-5.8.6.1.

       The compressor MUST set both occurrences of the CC field to the
       same value.

       The decompressor MUST use the value of the CC field from the
       Encoding Type 0 within the Generic CRSC list, and it MUST thus
       ignore the first occurrence of the CC field."

5.3.  Bit Masks in List Compression

   The insertion and/or removal schemes, described in RFC 3095-Sections
   5.8.6.2 - 5.8.6.4, use bit masks to indicates insertion or removal
   positions within the reference list.  The size of the bit mask can be
   7 bits or 15 bits.

   The compressor MAY use a 7-bit mask, even if the reference list has
   more than seven items, provided that changes to the list are only
   applied to items within the first seven items of the reference list,
   leaving items with an index not covered by the 7-bit mask unchanged.
   The decompressor MUST NOT modify items with an index not covered by
   the 7-bit mask, when a 7-bit mask is received for a reference list
   that contains more than seven items.

5.4.  Headers Compressed with List Compression

   In RFC 3095-Section 5.8, it states that headers that can be part of
   extension header chains "include" AH [14], ESP NULL [13], minimal
   encapsulation (MINE) [15], GRE [16][17], and IPv6 [9] extensions.
   This list of headers that can be compressed is correct, but the word
   "include" should not be there, since only the header types listed can
   actually be handled.  It should further be noted that for the Minimal
   Encapsulation (MINE) header, there is no explicit discussion of how
   to compress it, as the header is sent either uncompressed or fully
   compressed away.

Jonsson, et al.             Standards Track                    [Page 16]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

5.5.  ESP NULL Header List Compression

   Due to the offset of the fields in the trailer part of the ESP
   header, a compressor MUST NOT compress packets containing more than
   one NULL ESP [13] header, unless the second-outermost header is
   treated as a regular ESP [12] header and the packets are compressed
   using profile 0x0003.

5.6.  Translation Tables and Indexes for IP Extension Headers

   RFC 3095-Section 5.8.4 describes how list indexes are associated to
   list items and how table lists are built for IP extension headers.
   The text incorrectly states that one index per type is used, since
   the same type can appear several times with different content in one
   single chain.

   In IP extension header list compression, an index is associated with
   each individual extension header of an extension header chain.  When
   there are multiple non-identical occurrences of the same extension
   type (Protocol Number) within a header chain, each MUST be given its
   own index.

   In the case where there are multiple identical occurrences of the
   same extension type, the compressor can associate them to the same
   index.  When the value of an item whose index occurs more than once
   in the list is updated, the compressor MUST send the value for each
   occurrence of that index in the list.

   When content of extension headers changes, an implementation can
   choose to either use a different index or update the existing one.
   Some extensions can be compressed away even when some fields change,
   as those changes can be conveyed to the decompressor implicitly (e.g.
   sequence numbers in extension headers that can be inferred from the
   RTP SN) or explicitly (e.g., as part of the 'IP extension header(s)'
   field in Extension 3).

   When there is more than one IP header, there is more than one list of
   extension headers, and a translation table is maintained for each
   list independently of one another.

5.7.  Reference List

   A list compressed using encoding type 1 (insertion), type 2
   (removal), or type 3 (removal/insertion) uses a coding scheme that is
   based on the use of a reference list in the context (identified as
   ref_id).

Jonsson, et al.             Standards Track                    [Page 17]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   While it could seem to be a fair choice to send a type 1 list when
   ref_id is an empty list, there is nothing gained in doing so with
   respect to using a type 0 list.  Sending a type 2 list when ref_id is
   an empty list would lead to a failure, while sending a type 3 list
   has very little meaning.  All these alternatives could be seen as
   possible, based on how list compression is specified in RFC 3095.

   If these alternatives were allowed, a decompressor would become
   required to maintain a sliding window of ref_id lists in R-mode, even
   for the case where no items are sent in the compressed list, and this
   is not a desirable requirement.  Using list encoding type 1, type 2,
   and type 3 is therefore only allowed for non-empty reference lists.

   FORMAL ADDITION TO RFC 3095:

      "Regardless of the operating mode, for list encoding of type 1,
       type 2, and type 3 lists, ref_id MUST refer to a non-empty list."

5.8.  Compression of AH and GRE Sequence Numbers

   RFC 3095-Section 5.8.4.2 and RFC 3095-Section 5.8.4.4 describe how to
   compress the Authentication Header (AH) [14] and the Generic Routing
   Encapsulation (GRE) [16][17] header.  Both these sections present a
   possibility to omit the AH/GRE sequence number in the compressed
   header, under certain circumstances.  However, the specific
   conditions for omitting the AH/GRE sequence number, as well as the
   concrete compression and decompression procedures to apply, are not
   clearly defined to guarantee robustness and facilitate interoperable
   implementation.

   Proper rules are provided for the ESP case, i.e.,:

      "Sequence Number: Not sent when the offset from the sequence
       number of the compressed header is constant, when the compressor
       has confidence that the decompressor has established the correct
       offset.  When the offset is not constant, the sequence number may
       be compressed by sending LSBs"

   The same logic applies to the AH/GRE sequence numbers.

   INCORRECT RFC 3095 TEXT (RFC 3095-Section 5.8.4.2):

      "If the sequence number in the AH linearly increases as the RTP
       Sequence Number increases, and the compressor is confident that
       the decompressor has obtained the pattern, the sequence number in
       AH need not be sent.  The decompressor applies linear
       extrapolation to reconstruct the sequence number in the AH."

Jonsson, et al.             Standards Track                    [Page 18]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   CORRECTED TEXT:

      "The AH sequence number can be omitted from the compressed header
       when the offset from the sequence number (SN) of the compressed
       header is constant, when the compressor has confidence that the
       decompressor has established the correct offset."

   INCORRECT RFC 3095 TEXT (RFC 3095-Section 5.8.4.4):

      "If the sequence number in the GRE header linearly increases as
       the RTP Sequence Number increases and the compressor is confident
       that the decompressor has received the pattern, the sequence
       number in GRE need not be sent.  The decompressor applies linear
       extrapolation to reconstruct the sequence number in the GRE
       header."

   CORRECTED TEXT:

      "The GRE sequence number can be omitted from the compressed header
       when the offset from the sequence number (SN) of the compressed
       header is constant, when the compressor has confidence that the
       decompressor has established the correct offset."

6.  Updating Properties

6.1.  Implicit Updates

   A context updating packet that contains compressed sequence number
   information may also carry information about other fields; in such
   cases, these fields are updated according to the content of the
   packet.  The updating packet also implicitly updates inferred fields
   (e.g., RTP Timestamp) according to the current mode and the
   appropriate mapping function of the updated and inferred fields.

   An updating packet thus updates the reference values of all header
   fields, either explicitly or implicitly, except for the UO-1-ID
   packet (see Section 6.2 of this document).  In UO-mode, all packets
   are updating packets, while in R-mode, all packets with a CRC are
   updating packets.

   For example, a UO-0 packet contains the compressed RTP sequence
   number (SN).  Such a packet also implicitly updates RTP timestamp,
   IPv4 ID, and sequence numbers of IP extension headers.

Jonsson, et al.             Standards Track                    [Page 19]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

6.2.  Updating Properties of UO-1*

   RFC 3095-Section 5.7.3 states that the values provided in extensions
   carried by a UO-1-ID packet do not update the context, except for SN,
   TS, or IP-ID fields.  However, RFC 3095-Section 5.8.1 correctly
   states that the translation table in the context is updated whenever
   an (Index, item) pair is received, something that is contradicted by
   the statement in RFC 3095-5.7.3 because the UO-1-ID packet can carry
   Extension 3 with (Index, item) pair items within the 'Compressed CSRC
   list' field.  In addition to this contradiction, the text does not
   mention what to do with the other sequence numbers inferred from the
   SN, which are also to be implicitly updated.  The updating properties
   of UO-1* as stated by RFC 3095-Section 5.7.3 are thus incomplete.

   INCOMPLETE RFC 3095 TEXT (RFC 3095-Section 5.7.3):

      "Values provided in extensions, except those in other SN, TS, or
       IP-ID fields, do not update the context."

   CORRECTED TEXT:

      "UO-1-ID packets only updates TS, SN, IP-ID, and sequence numbers
       of IP extension headers.  Other values provided in extensions do
       not update the context.

       The decompressor MUST update its translation table whenever an
       (Index, item) pair is received, as per RFC 3095-Section 5.8.1,
       and this rule applies also to UO-1-ID packets."

6.3.  Context Updating Properties for IR Packets

   IR packets do not clear the whole context, but update all fields
   carried in the IR header.  Similarly, an IR without a dynamic chain
   simply updates the static part of the context, while the rest of the
   context is left unchanged.

   A consequence of this is that fields that are not updated by the IR
   packet, e.g., the translation tables for list compression, MUST NOT
   be invalidated by the decompressor when it assumes context damage.

6.4.  RTP Padding Field (R-P) in Extension 3

   RFC 3095-Section 5.7.5 defines the properties of RTP header flags and
   fields in Extension 3.  These get updated when the rtp flag of the
   Extension 3 is set, i.e., when rtp = 1; otherwise, they are not
   updated.  However, it is unclear how Extension 3 updates the R-P bit
   in the context.

Jonsson, et al.             Standards Track                    [Page 20]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   INCOMPLETE RFC 3095 TEXT (RFC 3095-Section 5.7.5):

      "R-P: RTP Padding bit, absolute value (presumed zero if absent)."

   CORRECTED TEXT:

      "R-P: RTP Padding bit.  If R-PT = 1, R-P is the absolute value of
            the RTP padding bit and this value updates context(R-P).  If
            R-PT = 0, context(R-P) is updated to zero."

6.5.  RTP eXtension bit (X) in dynamic part

   RFC 3095-Section 5.7.7.6 defines the properties of the RTP header
   flags and fields in the RTP part of the dynamic chain of IR and IR-
   DYN packets.  However, it is unclear how the X bit is updated in the
   context.

   INCOMPLETE RFC 3095 TEXT (RFC 3095-Section 5.7.7.6):

      "X: Copy of X bit from RTP header (presumed 0 if RX = 0)"

   CORRECTED TEXT:

      "X: X bit from RTP header.  If RX = 1, X is the X bit from the RTP
          header and this value updates context(X).  If RX = 0,
          context(X) is updated to zero."

7.  Context management and CID/context Reuse

7.1.  Persistence of Decompressor Contexts

   As part of the negotiated channel parameters, compressor and
   decompressor have, through the MAX_CID parameter, agreed on the
   highest context identification (CID) number to be used.  By agreeing
   on MAX_CID, the decompressor also agrees to provide memory resources
   to host at least MAX_CID+1 contexts, and an established context with
   a CID within this negotiated space MUST be kept by the decompressor
   until either the CID gets reused, or the channel is taken down or
   renegotiated.

7.2.  CID/Context Reuse

   As part of the channel negotiation, the maximal number of active
   contexts supported is negotiated between the compressor and the
   decompressor through the MAX_CID parameter.  The value of MAX_CID can
   differ significantly from one link application to another, as well as
   the load in terms of the number of packet streams to compress.  The
   lifetime of a ROHC channel can also vary, from almost permanent to

Jonsson, et al.             Standards Track                    [Page 21]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   rather short-lived.  However, in general, it is not expected that
   resources will be allocated for more contexts than what can
   reasonably be expected to be active concurrently over the link.  As a
   consequence hereof, context identifiers (CIDs) and context memory are
   resources that will have to be reused by the compressor as part of
   what can be considered normal operation.

   How context resources are reused is left unspecified in RFC 3095 [1]
   and subsequent 3095-based ROHC specifications.  This document does
   not intend to change that, i.e., ROHC resource management is still
   considered an implementation detail.  However, reusing a CID and its
   allocated memory is not always as simple as initiating a context with
   a previously unused CID.  Because some profiles can be operating in
   various modes where packet formats vary depending on current mode,
   care has to be taken to ensure that the old context data will be
   completely and safely overwritten, eliminating the risk of undesired
   side effects from interactions between old and new context data.
   This document therefore points out some important core aspects to
   consider when implementing resource management in ROHC compressors
   and decompressors.

   On a high level, CID/context reuse can be of two kinds, either reuse
   for a new context based on the same profile as the old context, or
   for a new context based on a different profile.  These cases are
   discussed separately in the following two sub-sections.

7.2.1.  Reusing a CID/Context with the Same Profile

   For multi-mode profiles, such as those defined in RFC 3095 [1], mode
   transitions are performed using a decompressor-initiated handshake
   procedure, as defined in RFC 3095-Section 5.6.  When a CID/context is
   reused for a new context based on the same profile as the old
   context, the current mode of operation SHOULD be inherited from the
   old to the new context.  Specifically, the compressor SHOULD continue
   to operate using the mode of operation of the old context also with
   the new context.  The reason for this is that there is no reliable
   way for the compressor to inform the decompressor that a CID/context
   reuse is happening.  The decompressor can thus not be expected to
   clear the context memory for the CID (see Section 6.3), and there is
   no way to trigger a safe mode switching (which requires the
   decompressor-initiated handshake procedure).

   The rule of mode inheritance applies also when the
   CONTEXT_REINITIALIZATION signal (RFC 3095-Section 6.3.1) is used to
   reinitiate an entire context.

Jonsson, et al.             Standards Track                    [Page 22]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

7.2.2.  Reusing a CID/Context with a Different Profile

   When a CID is reused for a new context based on a different profile
   than the old context, both the compressor and the decompressor MUST
   start operation with that context in the initial mode of the profile
   (if it is a multi-mode profile).  This applies both to IR-initiated
   new contexts and profile downgrades with IR-DYN (e.g., the profile
   0x0001 -> profile 0x0002 downgrade in RFC 3095-Section 5.11.1).

   Type 0 and type 1 packets have different formats in U/O- and R-mode,
   and these R-mode packets have no CRC.  When initiating a new context
   on a reused R-mode CID, there is a risk that the decompressor will
   misinterpret compressed packets if the initiating IR packets are
   lost.

   A CID for a context currently operating in R-mode SHOULD therefore
   not be reused for a new context based on a different profile than the
   old context.  A compressor doing otherwise should minimize the risk
   for misinterpretation of R-0/R-1 by, e.g., not using packets of types
   beginning with 00 or 10 before it is highly confident that the new
   context has successfully been initiated at the decompressor.

8.  Other Protocol Clarifications

8.1.  Meaning of NBO

   In IPv4 dynamic part (RFC 3095-Section 5.7.7.4), if the 'NBO' bit is
   set, it means that network byte order is used.

8.2.  IP-ID

   According to RFC 3095-Section 5.7, IP-ID means the compressed value
   of the IPv4 header's 'Identification' field.  Compressed packets
   contain this compressed value (IP-ID), while IR packets with dynamic
   chain and IR-DYN packets transmit the original, uncompressed
   Identification field value.  The IP-ID field always represents the
   Identification value of the innermost IPv4 header whose corresponding
   RND flag is not 1.

   If RND or RND2 is set to 1, the corresponding IP-ID(s) is (are) sent
   as 16-bit uncompressed Identification value(s) at the end of the
   compressed base header, according to the IP-ID description (see the
   beginning of RFC 3095-Section 5.7).  When there is no compressed IP-
   ID, i.e., for IPv6 or when all IP Identification information is sent
   as is (as indicated by RND/RND2 being set to 1), the decompressor
   ignores IP-ID bits sent within compressed base headers.

Jonsson, et al.             Standards Track                    [Page 23]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   When RND=RND2=0, IP-ID is compressed, i.e., expressed as an SN offset
   and byte-swapped if NBO=0.  This is the case also when 16 bits of
   IP-ID is sent in Extension 3.

   When RND=0 but no IP-ID bits are sent in the compressed header, the
   SN offset for IP-ID stays unchanged, meaning that Offset_m equals
   Offset_ref, as described in Section 4.5.5.  This is further expressed
   in a slightly different way (with the same meaning) in Section 5.7,
   where it is said that "default-slope(IP-ID offset) = 0", meaning, if
   no bits are sent for IP-ID, its SN offset slope defaults to 0.

8.3.  Extension-3 in UOR-2* Packets

   Some flags of the IP header in the extension (e.g., NBO or RND) may
   change the interpretation of fields in UOR-2* packets.  In such
   cases, when a flag changes in Extension 3, a decompressor MUST re-
   parse the UOR-2* packet.

8.4.  Multiple Occurrences of the M Bit

   The RTP header part of Extension 3, as defined by RFC 3095-Section
   5.7.5, includes a one-bit field for the RTP Marker bit.  This field
   is also present in all compressed base header formats except for UO-
   1-ID; meaning, there may be two occurrences of the field within one
   single compressed header.  In such cases, the two M fields must have
   the same value.

   FORMAL ADDITION TO RFC 3095:

      "When there are two occurrences of the M field in a compressed
       header (both in the compressed base header and in the RTP part of
       Extension 3), the compressor MUST set both these occurrences of
       the M field to the same value.

       At the decompressor, if the two M field values of such a packet
       are not identical, the packet MUST be discarded."

8.5.  Multiple SN options in one feedback packet

   The length of the sequence number field in the original ESP [12]
   header is 32 bits.  The format of the SN feedback option (RFC 3095-
   Section 5.7.6.6) allows for 8 additional SN bits to the 12 SN bits of
   the FEEDBACK-2 format (RFC 3095-Section 5.7.6.1).  One single SN
   feedback option is thus not enough for the decompressor to send back
   all the 32 bits of the ESP sequence number in a feedback packet,
   unless it uses multiple SN options in one feedback packet.

Jonsson, et al.             Standards Track                    [Page 24]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   RFC 3095-Section 5.7.6.1 declares that a FEEDBACK-2 packet can
   contain a variable number of feedback options, and the options can
   appear in any order.

   When processing multiple SN options in one feedback packet, the SN
   would be given by concatenating the fields.

8.6.  Multiple CRC Options in One Feedback Packet

   Although it is not useful to have more than one single CRC option in
   a feedback packet, having multiple CRC options is still allowed.  If
   multiple CRC options are included, all such CRC options MUST be
   identical, as they will be calculated over the same header; the
   compressor MUST otherwise discard the feedback packet.

8.7.  Responding to Lost Feedback Links

   Although this is neither desirable or expected, it may happen that a
   link used to carry feedback between two associated instances becomes
   unavailable.  If the compressor can be notified of such an event, the
   compressor SHOULD restart compression for each flow that is operating
   in R-mode.  When restarting compression, the compressor SHOULD use a
   different CID for each flow being restarted; this is useful to avoid
   the possibility of misinterpreting the type of the compressed header
   for the packet type identifiers that are common to both U/O-mode and
   R-mode, when the flow is restarted in U-mode (see also Section 7.2).

   Generally, feedback links are not expected to disappear once present,
   but it should be noted that this might be the case for certain link
   technologies.

8.8.  UOR-2 in Profile 0x0002 (UDP) and Profile 0x0003 (ESP)

   One single new format is defined for UOR-2 in profile 0x0002 and
   profile 0x0003, which replaces all three (UOR-2, UOR-2-ID, UOR-2-TS)
   formats from profile 0x0001.  The same UOR-2 format is thus used
   independent of whether or not there are IP headers with a
   corresponding RND=1.  This also applies to the IP profile [4] and the
   IP/UDP-Lite profile [5].

8.9.  Sequence Number LSB's in IP Extension Headers

   In RFC 3095-Section 5.8.5, formats are defined for compression of IP
   extension header fields.  These include compressed sequence number
   fields, and these fields contain the "LSB of sequence number".  These
   sequence numbers are not "LSB-encoded" as, e.g., the RTP sequence
   number, but are the LSB's of the uncompressed fields.

Jonsson, et al.             Standards Track                    [Page 25]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

8.10.  Expecting UOR-2 ACKs in O-Mode

   Usage of UOR-2 ACKs in O-mode, as discussed in RFC 3095-Section
   5.4.1.1.2, is optional.  A decompressor can also send ACKs for
   purposes other than to acknowledge the UOR-2, without having to
   continue sending ACKs for all UOR-2.  Similarly, a compressor
   implementation can ignore UOR-2s ACKs for the purpose of adapting the
   optimistic approach strategies.

   It is thus NOT RECOMMENDED to use the optional ACK mechanism in O-
   mode, either in compressor or in decompressor implementations.

   Using an incorrect expectation on UOR-2 ACKs as a basis for
   compressor behavior will significantly degrade the compression
   performance.  This is because UOR-2 ACKs can be sent from a
   decompressor for other purposes than to acknowledge the UOR-2 packet,
   e.g., to send feedback such as clock resolution, or to initiate a
   mode transition.  If an implementation does use the optional
   acknowledgment algorithm described in Section 5.4.1.1.2, it must make
   sure to set the k_3 and n_3 parameters to much larger values than 1
   to ensure that the compressor performance is not degraded due to the
   problem described above.

8.11.  Context Repairs, TS_STRIDE and TIME_STRIDE

   The 7-bit CRC used to verify the outcome of the decompression attempt
   covers the original uncompressed header.  The CRC verification thus
   excludes TS_STRIDE and TIME_STRIDE, as these fields are not part of
   the original uncompressed header.

   The UOR-2 packet type can be used to update the value of the
   TS_STRIDE and/or the TIME_STRIDE, with the Extension 3.  However,
   these fields are not used for decompression of the RTP TS field for
   this packet type and their respective value is thus not verified,
   either implicitly or explicitly.

   When the compressor receives a negative acknowledgement, it thus
   cannot determine whether the failure may be caused by an unsuccessful
   update to the TS_STRIDE and/or the TIME_STRIDE field(s), for which a
   previous header that last attempted to update their value had
   previously been acknowledged.

   FORMAL ADDITION TO RFC 3095:

      "When the compressor receives a NACK and uses the UOR-2 header
       type to repair the decompressor context, it SHOULD include fields
       that update the value of both the TS_STRIDE and the TIME_STRIDE
       whose value it has updated at least once since the establishment

Jonsson, et al.             Standards Track                    [Page 26]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

       of that context, i.e., since the CID was first associated with
       its current profile.

       When the compressor receives a static-NACK, it MUST include in
       the IR header fields for both the TS_STRIDE and the TIME_STRIDE
       whose value it has updated at least once since the establishment
       of that context, i.e., since the CID was first associated with
       its current profile."

9.  ROHC Negotiation

   RFC 3095-Section 4.1 states that the link layer must provide means to
   negotiate, e.g., the channel parameters listed in RFC 3095-Section
   5.1.1.  One of these parameters is the PROFILES parameter, which is a
   set of non-negative integers where each integer indicates a profile
   supported by the decompressor.

   Each profile is identified by a 16-bit value, where the 8 LSB bits
   indicate the actual profile, and the 8 MSB bits indicate the variant
   of that profile (see RFC 3095-Section 8).  In the ROHC headers sent
   over the link, the profile used is identified only with the 8 LSB
   bits, which means that the compressor and decompressor must have
   agreed on which variant to use for each profile.

   The negotiation protocol must thus be able to communicate to the
   compressor the set of profiles supported by the decompressor.  When
   multiple variants of the same profile are available, the negotiation
   protocol must provide the means for the decompressor to know which
   variant will be used by the compressor.  This basically means that
   the PROFILES set after negotiation MUST NOT include more than one
   variant of a profile.

10.  PROFILES Sub-option in ROHC-over-PPP

   The logical union of sub-options for IPCP and IPV6CP negotiations, as
   specified by ROHC over PPP [2], cannot be used for the PROFILES
   suboption, as the whole union would then have to be considered within
   each of the two IPCP negotiations to avoid getting an ambiguous
   profile set.  An implementation of RFC 3241 MUST therefore ensure
   that the same profile set is negotiated for both IPv4 and IPv6
   (IPCP/IPV6CP).

11.  Constant IP-ID Encoding in IP-only and UPD-Lite Profiles

   In the ROHC IP-only profile, Section 3.3 of RFC 3843 [4], a mechanism
   for encoding of a constant Identification value in IPv4 (constant
   IP-ID) is defined.  This mechanism is also used by the ROHC UDP-Lite
   profiles, RFC 4019 [5].

Jonsson, et al.             Standards Track                    [Page 27]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

   The "Constant IP-ID" mechanism applies to both the inner and outer IP
   header, when present, meaning that there will be both a SID and a
   SID2 context value.

12.  Security Considerations

   This document provides a number of corrections and clarifications to
   [1], but it does not make any changes with regard to the security
   aspects of the protocol.  As a consequence, the security
   considerations of [1] apply without additions.

13.  Acknowledgments

   The authors would like to thank Vicknesan Ayadurai, Carsten Bormann,
   Mikael Degermark, Zhigang Liu, Abigail Surtees, Mark West, Tommy
   Lundemo, Alan Kennington, Remi Pelland, Lajos Zaccomer, Endre Szalai,
   Mark Kalmanczhelyi, and Arpad Szakacs for their contributions and
   comments.  Thanks also to the committed document reviewers, Carl
   Knutsson and Biplab Sarkar, who reviewed the document during working
   group last-call.

14.  References

14.1.  Normative References

   [1]  Bormann, C., Burmeister, C., Degermark, M., Fukushima, H.,
        Hannu, H., Jonsson, L-E., Hakenberg, R., Koren, T., Le, K., Liu,
        Z., Martensson, A., Miyazaki, A., Svanbro, K., Wiebke, T.,
        Yoshimura, T., and H. Zheng, "RObust Header Compression (ROHC):
        Framework and four profiles: RTP, UDP, ESP, and uncompressed",
        RFC 3095, July 2001.

   [2]  Bormann, C., "Robust Header Compression (ROHC) over PPP", RFC
        3241, April 2002.

   [3]  Simpson, W., "PPP in HDLC-like Framing", STD 51, RFC 1662, July
        1994.

   [4]  Jonsson, L-E. and G. Pelletier, "RObust Header Compression
        (ROHC): A Compression Profile for IP", RFC 3843, June 2004.

   [5]  Pelletier, G., "RObust Header Compression (ROHC): Profiles for
        User Datagram Protocol (UDP) Lite", RFC 4019, April 2005.

   [6]  Bradner, S., "Key words for use in RFCs to Indicate Requirement
        Levels", BCP 14, RFC 2119, March 1997.

Jonsson, et al.             Standards Track                    [Page 28]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

14.2.  Informative References

   [7]  Jonsson, L-E., Pelletier, G., and K. Sandlund, "RObust Header
        Compression (ROHC): A Link-Layer Assisted Profile for
        IP/UDP/RTP", RFC 4362, January 2006.

   [8]  Postel, J., "Internet Protocol", STD 5, RFC 791, September 1981.

   [9]  Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6)
        Specification", RFC 2460, December 1998.

   [10] Postel, J., "User Datagram Protocol", STD 6, RFC 768, August
        1980.

   [11] Schulzrinne, H., Casner, S., Frederick, R., and V. Jacobson,
        "RTP: A Transport Protocol for Real-Time Applications", STD 64,
        RFC 3550, July 2003.

   [12] Kent, S., "IP Encapsulating Security Payload (ESP)", RFC 4303,
        December 2005.

   [13] Glenn, R. and S. Kent, "The NULL Encryption Algorithm and Its
        Use With IPsec", RFC 2410, November 1998.

   [14] Kent, S., "IP Authentication Header", RFC 4302, December 2005.

   [15] Perkins, C., "Minimal Encapsulation within IP", RFC 2004,
        October 1996.

   [16] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. Traina,
        "Generic Routing Encapsulation (GRE)", RFC 2784, March 2000.

   [17] Dommety, G., "Key and Sequence Number Extensions to GRE", RFC
        2890, September 2000.

Jonsson, et al.             Standards Track                    [Page 29]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

Appendix A.  Sample CRC Algorithm

   #!/usr/bin/perl -w
   use strict;
   #=================================
   #
   # ROHC CRC demo - Carsten Bormann cabo@tzi.org 2001-08-02
   #
   # This little demo shows the four types of CRCs in use in RFC 3095,
   # the specification for robust header compression. Type your data in
   # hexadecimal form and then press Control+D.
   #
   #---------------------------------
   #
   # utility
   #
   sub dump_bytes($) {
       my $x = shift;
       my $i;
       for ($i = 0; $i < length($x); ) {
     printf("%02x ", ord(substr($x, $i, 1)));
     printf("\n") if (++$i % 16 == 0);
       }
       printf("\n") if ($i % 16 != 0);
   }

   #---------------------------------
   #
   # The CRC calculation algorithm.
   #
   sub do_crc($$$) {
       my $nbits = shift;
       my $poly = shift;
       my $string = shift;

       my $crc = ($nbits == 32 ? 0xffffffff : (1 << $nbits) - 1);
       for (my $i = 0; $i < length($string); ++$i) {
         my $byte = ord(substr($string, $i, 1));
         for( my $b = 0; $b < 8; $b++ ) {
           if (($crc & 1) ^ ($byte & 1)) {
             $crc >>= 1;
             $crc ^= $poly;
           } else {
           $crc >>= 1;
           }
           $byte >>= 1;
         }
       }

Jonsson, et al.             Standards Track                    [Page 30]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

       printf "%2d bits, ", $nbits;
       printf "CRC: %02x\n", $crc;
   }

   #---------------------------------
   #
   # Test harness
   #
   $/ = undef;
   $_ = <>;         # read until EOF
   my $string = ""; # extract all that looks hex:
   s/([0-9a-fA-F][0-9a-fA-F])/$string .= chr(hex($1)), ""/eg;
   dump_bytes($string);

   #---------------------------------
   #
   # 32-bit segmentation CRC
   # Note that the text implies that this is complemented like for PPP
   # (this differs from 8-, 7-, and 3-bit CRCs)
   #
   #      C(x) = x^0 + x^1 + x^2 + x^4 + x^5 + x^7 + x^8 + x^10 +
   #             x^11 + x^12 + x^16 + x^22 + x^23 + x^26 + x^32
   #
   do_crc(32, 0xedb88320, $string);

   #---------------------------------
   #
   # 8-bit IR/IR-DYN CRC
   #
   #      C(x) = x^0 + x^1 + x^2 + x^8
   #
   do_crc(8, 0xe0, $string);

   #---------------------------------
   #
   # 7-bit FO/SO CRC
   #
   #      C(x) = x^0 + x^1 + x^2 + x^3 + x^6 + x^7
   #
   do_crc(7, 0x79, $string);

   #---------------------------------
   #
   # 3-bit FO/SO CRC
   #
   #      C(x) = x^0 + x^1 + x^3
   #
   do_crc(3, 0x6, $string);

Jonsson, et al.             Standards Track                    [Page 31]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

Authors' Addresses

   Lars-Erik Jonsson
   Optand 737
   SE-831 92 Ostersund, Sweden
   Phone: +46 70 365 20 58
   EMail: lars-erik@lejonsson.com

   Kristofer Sandlund
   Ericsson AB
   Box 920
   SE-971 28 Lulea, Sweden
   Phone: +46 8 404 41 58
   EMail: kristofer.sandlund@ericsson.com

   Ghyslain Pelletier
   Ericsson AB
   Box 920
   SE-971 28 Lulea, Sweden
   Phone: +46 8 404 29 43
   EMail: ghyslain.pelletier@ericsson.com

   Peter Kremer
   Conformance and Software Test Laboratory
   Ericsson Hungary
   H-1300 Bp. 3., P.O. Box 107, HUNGARY
   Phone: +36 1 437 7033
   EMail: peter.kremer@ericsson.com

Jonsson, et al.             Standards Track                    [Page 32]



RFC 4815       Corrections and Clarifications to RFC 3095  February 2007

Full Copyright Statement

   Copyright (C) The IETF Trust (2007).

   This document is subject to the rights, licenses and restrictions
   contained in BCP 78, and except as set forth therein, the authors
   retain all their rights.

   This document and the information contained herein are provided on an
   "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
   OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND
   THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS
   OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
   THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
   WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Intellectual Property

   The IETF takes no position regarding the validity or scope of any
   Intellectual Property Rights or other rights that might be claimed to
   pertain to the implementation or use of the technology described in
   this document or the extent to which any license under such rights
   might or might not be available; nor does it represent that it has
   made any independent effort to identify any such rights.  Information
   on the procedures with respect to rights in RFC documents can be
   found in BCP 78 and BCP 79.

   Copies of IPR disclosures made to the IETF Secretariat and any
   assurances of licenses to be made available, or the result of an
   attempt made to obtain a general license or permission for the use of
   such proprietary rights by implementers or users of this
   specification can be obtained from the IETF on-line IPR repository at
   http://www.ietf.org/ipr.

   The IETF invites any interested party to bring to its attention any
   copyrights, patents or patent applications, or other proprietary
   rights that may cover technology that may be required to implement
   this standard.  Please address the information to the IETF at
   ietf-ipr@ietf.org.

Acknowledgement

   Funding for the RFC Editor function is currently provided by the
   Internet Society.

Jonsson, et al.             Standards Track                    [Page 33]