Remote checksum offload for encapsulation
draft-herbert-remotecsumoffload-00

INTERNET-DRAFT                                                T. Herbert
Intended Status: Informational                                    Google
Expires: February 2015                                   August 27, 2014

               Remote checksum offload for encapsulation
                   draft-herbert-remotecsumoffload-00

Status of this Memo

   This Internet-Draft is submitted to IETF in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as
   Internet-Drafts.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   The list of current Internet-Drafts can be accessed at
   http://www.ietf.org/1id-abstracts.html

   The list of Internet-Draft Shadow Directories can be accessed at
   http://www.ietf.org/shadow.html

Copyright and License Notice

   Copyright (c) 2014 IETF Trust and the persons identified as the
   document authors. All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document. Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document. Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

 

Herbert                  Expires February 2015                  [Page 1]
INTERNET DRAFT     draft-herbert-remotecsumoffload-00    August 27, 2014

Abstract

   This specification describes remote checksum offload for
   encapsulation, which is a mechanism that provides checksum offload of
   encapsulated packets using rudimentary offload capabilities found in
   most Network Interface Card (NIC) devices. The outer header checksum
   (e.g. that in UDP or GRE) is enabled in packets and, with some
   additional meta information, a receiver is able to deduce the
   checksum to be set for an inner encapsulated packet. Effectively this
   offloads the computation of the inner checksum. Enabling the outer
   checksum in encapsulation has the additional advantage that it covers
   more of the packet than the inner checksum including the
   encapsulation headers.

Table of Contents

   1 Introduction  . . . . . . . . . . . . . . . . . . . . . . . . . . 3
   2 Checksum offload background . . . . . . . . . . . . . . . . . . . 3
      2.1 The Internet checksum  . . . . . . . . . . . . . . . . . . . 3
      2.2 Transmit checksum offload  . . . . . . . . . . . . . . . . . 4
         2.2.1 Generic transmit offload  . . . . . . . . . . . . . . . 4
         2.2.2 Protocol specific transmit offload  . . . . . . . . . . 4
      2.3 Receive checksum offload . . . . . . . . . . . . . . . . . . 5
         2.3.1 CHECKSUM_COMPLETE . . . . . . . . . . . . . . . . . . . 5
         2.3.2 CHECKSUM_UNNECESSARY  . . . . . . . . . . . . . . . . . 5
   3 Remote checksum offload . . . . . . . . . . . . . . . . . . . . . 5
      3.1 Meta data format . . . . . . . . . . . . . . . . . . . . . . 6
      3.2 Transmit operation . . . . . . . . . . . . . . . . . . . . . 6
      3.3 Receiver operation . . . . . . . . . . . . . . . . . . . . . 7
      3.4 Interaction with TCP segmentation offload  . . . . . . . . . 8
   4  Security Considerations  . . . . . . . . . . . . . . . . . . . . 8
   5  IANA Considerations  . . . . . . . . . . . . . . . . . . . . . . 8
   6  References . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
      6.1  Normative References  . . . . . . . . . . . . . . . . . . . 8
      6.2  Informative References  . . . . . . . . . . . . . . . . . . 9
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . . . 9

 

Herbert                  Expires February 2015                  [Page 2]
INTERNET DRAFT     draft-herbert-remotecsumoffload-00    August 27, 2014

1 Introduction

   Checksum offload is a capability of NICs where the checksum
   calculation for a transport layer packet (TCP, UDP, etc.) is
   performed by a device on behalf of the host stack. Checksum offload
   is applicable to both transmit and receive, where on transmit the
   device writes the computed checksum into the packet, and on receive
   the device provides the computed checksum of the packet or an
   indication that specific transport checksums were validated. This
   feature saves CPU cycles in the host and has become ubiquitous in
   modern NICs. 

   A host may both source transport packets and encapsulate them for
   transit over an underlying network. In this case, checksum offload is
   still desirable, but now must be done on an encapsulated packet. Many
   deployed NICs are only capable of providing checksum offload for
   simple TCP or UDP packets. Such NICs typically use protocol specific
   mechanisms where they must parse headers in order to perform checksum
   calculations. Updating these NICs to perform checksum offload for
   encapsulation requires new parsing logic which is likely infeasible
   or at cost prohibitive.

   In this specification we describe an alternative that uses
   rudimentary NIC offload features to support offloading checksum
   calculation of encapsulated packets. In this design, the outer
   checksum is enabled on transmit, and meta information indicating the
   location of the checksum field being offloaded and its starting point
   for computation are sent with a packet. On receipt, after the outer
   checksum is verified, the receiver sets the offloaded checksum field
   per the computed packet checksum and the meta data.

2 Checksum offload background

   In this section we provide some background into checksum offload
   operation.

2.1 The Internet checksum

   The Internet checksum [RFC0791] is used by several Internet protocols
   including IP [RFC1122], TCP [RFC0793], UDP [RFC0768] and GRE
   [RFC2784]. Efficient checksum calculation is critical to good
   performance [RFC1071], and the mathematical properties are useful in
   incrementally updating checksums [RFC1624]. An early approach to
   implementing checksum offload in hardware is described in [RFC1936].

   TCP and UDP checksums cover a pseudo header which is composed of the
   source and destination addresses of the corresponding IP packet,
 

Herbert                  Expires February 2015                  [Page 3]
INTERNET DRAFT     draft-herbert-remotecsumoffload-00    August 27, 2014

   upper layer packet length, and protocol. The checksum pseudo header
   is defined in [RFC0768] and [RFC0793] for IPv4, and in [RFC2460] for
   IPv6.

2.2 Transmit checksum offload

   In transmit checksum offload, a host networking stack defers the
   calculation and setting of a transport checksum in the packet to the
   device. A device may provide checksum offload only for specific
   protocols, or may provide a generic interface. In either case, only
   one offloaded checksum per packet is typical.

   When using transmit checksum offload, a host stack must initialize
   the checksum field in the packet. This is done by setting to zero
   (GRE) or to the bitwise not of the pseudo header (UDP or TCP). The
   device proceeds by computing the packet checksum from the start of
   the transport header through to the end of the packet. The bitwise
   not of the resulting value is written in the checksum field of the
   transport packet.

2.2.1 Generic transmit offload

   A device can provide a generic interface for transmit checksum
   offload. Checksum offload is enabled by setting two fields in the
   transmit descriptor for a packet: start offset and checksum offset.
   The start offset indicates the byte in the packet where the checksum
   calculation should start. The checksum offset indicates the offset in
   the packet where the checksum value is to be written.

   The generic interface is protocol agnostic, however only supports one
   offloaded checksum per packet. It is conceivable that a NIC could
   provide offload for more checksums by defining more than one 
   checksum start, checksum offset pair in the transmit descriptor.

2.2.2 Protocol specific transmit offload

   Some devices support transmit checksum offload for very specific
   protocols. For instance, many legacy devices can only perform
   checksum offload for UDP/IP and TCP/IP packets. These devices parse
   transmitted packets in order to determine the checksum start and
   checksum offset. They may also ignore the value in the checksum field
   by setting it to zero for checksum computation and computing the
   checksum of the pseudo header themselves.

   Protocol specific transmit offload is limited to the protocols a
   device supports. To support checksum offload of an encapsulated
   packet, a device must be a able to parse the encapsulation layer in
   order to locate the inner packet. 
 

Herbert                  Expires February 2015                  [Page 4]
INTERNET DRAFT     draft-herbert-remotecsumoffload-00    August 27, 2014

2.3 Receive checksum offload

   Upon receiving a packet, a device may perform a checksum calculation
   over the packet or part of the packet depending on the protocol. A
   result of this calculation is returned in the meta data of the
   receive descriptor for the packet. The host stack can apply the
   result in verifying checksums as it processes the packet. The intent
   is that the offload will obviate the need for the networking stack to
   perform its own checksum calculation over the packet.

   There are two basic methods of receive checksum offload:
   CHECKSUM_COMPLETE and CHECKSUM_UNNECESSARY.

2.3.1 CHECKSUM_COMPLETE

   A device may calculate the checksum of a whole packet (layer 2
   payload) and return the resultant value to the host stack. The host
   stack can subsequently use this value to validate checksums in the
   packet. As the packet is parsed through various layers, the
   calculated checksum is updated to correspond to each layer (subtract
   out checksum for preceding bytes for a given header).

   CHECKSUM_COMPLETE is protocol agnostic and does not require any
   protocol awareness in the device. It works for any encapsulation and
   supports an arbitrary number of checksums in the packet.

2.3.2 CHECKSUM_UNNECESSARY

   A device may explicitly validate a checksum in a packet and return a
   flag in the receive descriptor that a transport checksum has been
   verified (host performing checksum computation is unnecessary). Some
   devices may be capable of validating more than one checksum in the
   packet, in which case the device returns a count of the number
   verified. Typically, only a positive signal is returned, if the
   device was unable to validate a checksum it does not return any
   information and the host will generally perform its own checksum
   computation. If a device returns a count of validations, this must
   refer to consecutive checksums that are present and validated in a
   packet (checksums cannot be skipped).

   CHECKSUM_UNNECESSARY is protocol specific, for instance in the case
   of UDP or TCP a device needs to consider the pseudo header in
   checksum validation. To support checksum offload of an encapsulated
   packet, a device must be able to parse the encapsulation layer in
   order to locate the inner packet.

3 Remote checksum offload

 

Herbert                  Expires February 2015                  [Page 5]
INTERNET DRAFT     draft-herbert-remotecsumoffload-00    August 27, 2014

   This section describes the remote checksum offload mechanism. This is
   primarily useful with UDP based encapsulation where the UDP checksum
   is enabled (not set to zero on transmit). The same technique could be
   applied to GRE encapsulation where the GRE checksum is enabled.

3.1 Meta data format

   Remote checksum offload requires the sending of meta data with an
   encapsulated packet. This data is a pair of checksum start and
   checksum offset values. More than one offloaded checksum could be
   supported if multiple pairs are sent.

   The meta data format for remote checksum offload is:

    0                   1                   2                   3
    0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |        Checksum start         |       Checksum offset         |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   o Checksum start: starting offset for checksum computation relative
   to the start of the encapsulation header. This is typically the
   offset of a transport header (e.g. UDP or TCP).

   o Checksum offset: Offset relative to the start of the encapsulation
   header where the derived checksum value is to be written. This
   typically is the offset of the checksum field in the transport header
   (e.g. UDP or TCP).

   Support for remote checksum offload with specific encapsulation
   protocols is outside the scope of this document, however any
   encapsulation format that supports some reasonable form of optional
   meta data should be amenable. In Generic UDP Encapsulation [GUE] this
   would entail defining an optional field, in Geneve [GENEVE] a TLV
   would be defined, for NSH [NSH] the meta data can either be in a
   service header or within a TLV. In any scenario, what the offsets in
   the meta data are relative to must be unambiguous (for instance when
   used in NSH the offsets may be relative to the NSH header itself).

3.2 Transmit operation

   The typical actions to set remote checksum offload on transmit are:

   1) Transport layer creates a packet and indicates in internal packet
   meta data that checksum is to be offloaded to the NIC (normal
   transport layer processing for checksum offload). The checksum field
   is populated with the bitwise not of the checksum of the pseudo
   header or zero as appropriate.
 

Herbert                  Expires February 2015                  [Page 6]
INTERNET DRAFT     draft-herbert-remotecsumoffload-00    August 27, 2014

   2) Encapsulation layer adds its headers to the packet including the
   offload meta data. The start offset and checksum offset are set
   accordingly.

   3) Encapsulation layer arranges for checksum offload of the outer
   header checksum (e.g. UDP).

   4) Packet is sent to the NIC. The NIC will perform transmit checksum
   offload and set the checksum field in the outer header. The inner
   header and rest of the packet are transmitted without modification.

3.3 Receiver operation

   The typical actions a host receiver does to support remote checksum
   offload are:

   1) Receive packet and validate outer checksum following normal
   processing (e.g. validate non-zero UDP checksum).

   2) Deduce full checksum for the IP packet. This is directly provided
   if device returns the packet checksum in CHECKSUM_COMPLETE. If the
   device returned CHECKSUM_UNNECESSARY, then the complete checksum can
   be trivially derived as either zero (GRE) or the bitwise not of the
   outer pseudo header (UDP).

   3) From the packet checksum, subtract the checksum computed from the
   start of the packet (outer IP header) to the offset in the packet
   indicted by checksum start in the meta data. The result is the
   deduced checksum to set in the checksum field of the encapsulated
   transport packet.

   In pseudo code:

     csum: initialized to checksum computed from start (outer IP header) to
           the end of the packet
     start_of_packet: address of start of packet
     offset_of_encap_hdr: relative to start_of_packet
     csum_start: value from meta data
     checksum(start, len): function to compute checksum from start address
           for len bytes

     csum -= checksum(start_of_packet, offset_of_encap_hdr + csum_start) 

   4) Write the resultant checksum value into the packet at the offset
   provided by checksum offset in the meta data.

   In pseudo code:
 

Herbert                  Expires February 2015                  [Page 7]
INTERNET DRAFT     draft-herbert-remotecsumoffload-00    August 27, 2014

     csum_offset: offset of checksum field

     *(start_of_packet + offset_of_encap_hdr + csum_offset) = csum

   5) Checksum is verified at the transport layer using normal
   processing. This should not require any checksum computation over the
   packet since the complete checksum has already been provided.

3.4 Interaction with TCP segmentation offload

   Remote checksum offload may be useful with TCP Segmentation Offload
   (TSO) in order to avoid host checksum calculations at the receiver.
   This can be implemented on a transmitter as follows:

   1) Host stack prepares a large segment for transmission including
   adding of encapsulation headers and the remote checksum option which
   refers to the encapsulated transport checksum in the large segment.

   2) TSO is performed by the device taking encapsulation into account.
   The outer checksum is computed and written for each packet. The inner
   checksum is not computed, and the encapsulation header (including
   checksum meta data) is replicated for each packet.

   3) At the receiver remote checksum offload processing occurs as
   normal for each packet.

4  Security Considerations

   Remote checksum offload should not impact protocol security.

5  IANA Considerations

   There are no IANA considerations in this specification. The remote
   checksum offload meta data may require an option number or type in
   specific encapsulation formats that support it.

6  References

6.1  Normative References

   [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, September
              1981.

   [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts -
              Communication Layers", STD 3, RFC 1122, October 1989.

   [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, RFC
 

Herbert                  Expires February 2015                  [Page 8]
INTERNET DRAFT     draft-herbert-remotecsumoffload-00    August 27, 2014

              793, September 1981.

   [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768,
              August 1980.

   [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. Traina,
              "Generic Routing Encapsulation (GRE)", RFC 2784, March
              2000.

   [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6
              (IPv6) Specification", RFC 2460, December 1998.

6.2  Informative References

   [RFC1071] Braden, R., Borman, D., and C. Partridge, "Computing the
              Internet checksum", RFC1071, September 1988.

   [RFC1624] Rijsinghani, A., Ed., "Computation of the Internet Checksum
              via Incremental Update", RFC1624, May 1994.

   [RFC1936] Touch, J. and B. Parham, "Implementing the Internet
              Checksum in Hardware", RFC1936, April 1996.

   [GUE]     Generic UDP Encapsulation draft-herbert-gue-01

   [GENEVE]  Geneve: Generic Network Virtualization Encapsulation draft-
              gross-geneve-01

   [NSH]     Network Service Header draft-quinn-sfc-nsh-03

Authors' Addresses

   Tom Herbert
   Google
   1600 Amphitheatre Parkway
   Mountain View, CA
   EMail: therbert@google.com

Herbert                  Expires February 2015                  [Page 9]