BIND 10 #1956: Implement perfdhcp IPv6 packet handling

[BIND 10 Development]

#1956: Implement perfdhcp IPv6 packet handling
-------------------------------------+-------------------------------------
                   Reporter:         |                 Owner:  stephen
  stephen                            |                Status:  reviewing
                       Type:         |             Milestone:  DHCP-
  enhancement                        |  Sprint-20120611
                   Priority:         |            Resolution:
  medium                             |             Sensitive:  0
                  Component:         |           Sub-Project:  DHCP
  perfdhcp                           |  Estimated Difficulty:  40
                   Keywords:         |           Total Hours:  39.5
            Defect Severity:  N/A    |
Feature Depending on Ticket:         |
        Add Hours to Ticket:  2      |
                  Internal?:  0      |
-------------------------------------+-------------------------------------
Changes (by marcin):

 * hours:  37.5 => 2
 * owner:  marcin => stephen
 * totalhours:  37.5 => 39.5


Comment:

 I implemented changes suggested in this review and committed to trac1955
 as this branch is now holding overall changes for #1955 and #1956.

 '''tests/tools/perfdhcp/perf_pkt6.cc'''
 I removed all invocations to bufferOut_.clear() and instead I used
 writeUintXAt() functions that allowed me to specify location where I
 wanted to write. I couldn't do the other way around because when I read
 options from the collection there is no guarantee that their offsets are
 in incremental order.

 Since we are operating on raw buffers and we don't parse whole the packet
 content we keep light restrictions on transaction id offset. It should be
 positive value and it must not be higher than packet size - 3 for DHCPv6
 and packet size - 4 for DHCPv4.

 '''tests/tools/perfdhcp/perf_pkt6.h'''
 In a function where there are multiple integer parameters inline forcing
 explicit object types is a good method to avoid user errors like putting
 things in wrong order. However in this particular case I tend to agree
 that this approach would produce inconsistency with other functions and
 parameters so I got rid of this.

 Please review branch trac1955.

-- 
Ticket URL: <http://bind10.isc.org/ticket/1956#comment:8>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development