GitLab

I suspect the previous version could cause the same type of use-after-free
bug, although I couldn't come up with an effective test scenario.

to enforce this style of creation, introduce a static factory method and
hide the constructor as private.

... now that we handle this with an offset_ptr inside setNamedAddress()
itself.

This commit introduces a problem that allMemoryDeallocated() doesn't
work anymore, because the reserved memory is freed only when the segment
is destroyed.

One workaround is to temporarily release and re-reserve this memory in
allMemoryDeallocated(), but that would make it a non-const method.  I
don't see any other clean way of doing this.

We don't attempt to recover in this case.

Also change some error handling to simply asserts.

Initially I thought it was not possible to test this as
ZoneTableSegment::create() was used to create the zone table segment,
but it seems it's possible to do it this way using a mock class.

seems to be necessary for some OS.  also explicitly include the relevant
header file.

Conflicts:
	src/lib/dhcp/tests/pkt6_unittest.cc

It would be better if these were static (to save on space and also to
use them inside ZoneTableSegment::create()), but unfortunately we cannot
make it virtual then. Another option would be to pass a static string
from ZoneTableSegment::create() onto the ZoneTableSegment constructor
and make it a non-virtual base class method.