RFC-3442 classless routes and DHCPD

[Philip Prindeville]

Still working on the OpenWRT integration of ISC-DHCP(D) and was hoping for some guidance.

I was looking at:

https://kb.isc.org/docs/isc-dhcp-44-manual-pages-dhcp-options#DEFINING%20NEW%20OPTIONS

As well as RFC 3442, in particular:

Classless Route Option Format

   The code for this option is 121, and its minimum length is 5 bytes.
   This option can contain one or more static routes, each of which
   consists of a destination descriptor and the IP address of the router
   that should be used to reach that destination.

    Code Len Destination 1    Router 1
   +-----+---+----+-----+----+----+----+----+----+
   | 121 | n | d1 | ... | dN | r1 | r2 | r3 | r4 |
   +-----+---+----+-----+----+----+----+----+----+

    Destination 2       Router 2
   +----+-----+----+----+----+----+----+
   | d1 | ... | dN | r1 | r2 | r3 | r4 |
   +----+-----+----+----+----+----+----+

   In the above example, two static routes are specified.

   Destination descriptors describe the IP subnet number and subnet mask
   of a particular destination using a compact encoding.  This encoding
   consists of one octet describing the width of the subnet mask,
   followed by all the significant octets of the subnet number.

   The width of the subnet mask describes the number of one bits in the
   mask, so for example a subnet with a subnet number of 10.0.127.0 and
   a netmask of 255.255.255.0 would have a subnet mask width of 24.

   The significant portion of the subnet number is simply all of the
   octets of the subnet number where the corresponding octet in the
   subnet mask is non-zero.  The number of significant octets is the
   width of the subnet mask divided by eight, rounding up, as shown in
   the following table:

        Width of subnet mask     Number of significant octets
                     0                     0
                  1- 8                     1
                  9-16                     2
                 17-24                     3
                 25-32                     4

   The following table contains some examples of how various subnet
   number/mask combinations can be encoded:

   Subnet number   Subnet mask      Destination descriptor
   0               0                0
   10.0.0.0        255.0.0.0        8.10
   10.0.0.0        255.255.255.0    24.10.0.0
   10.17.0.0       255.255.0.0      16.10.17
   10.27.129.0     255.255.255.0    24.10.27.129
   10.229.0.128    255.255.255.128  25.10.229.0.128
   10.198.122.47   255.255.255.255  32.10.198.122.47

And how to reconcile these two.  Don’t know why the option needed to “compact” the dotted quad… It just seems to create more work for no apparent gain (other than a pittance of octets saved).

Does anyone have a simple/clever way to define a new option for RFC-3442?

Without the compaction, this would be trivial as:

option classless-ipv4-route code 121 = { unsigned integer 8, ip-address, ip-address }; 

But that’s not the case.

I could do:

option classless-ipv4-route code 121 = { unsigned integer 8, array of { unsigned integer 8 }, ip-address }; 

But then I lose the notational convenience of ip-address.

Anyone?

Thanks,

-Philip