[SECNET PATCH v2 00/19] Fix polypath asymmetric routing

[Ian Jackson]

Now rebased on top of the "obvious bugfixes" I just (re)posted, and
tested.  I guess I will push this to master soon.

In the list below,
  + means a new patch in v2
  * means a patch modified in v2

Ian Jackson (19):
    NOTES: tiny fix
    configure: rerun autogen.sh with autoconf 2.69-10
  + site: transport peers update: avoid nearly-trivial debug
  + site: Log about crossed MSG1 ignored only once
  + site: Log about crossed MSG1 with a higher priority
  * site: Rename setup_priority to our_name_later
    site: Break out we_have_priority
  * polypath asymmetric routing: Priority to mobile sites
    site: Make return value of transforms be an enum
    site: Change return value of decrypt_msg0
    Introduce transform_apply_return_badseq
    Introduce transform_apply_seqdupe
    polypath asymmetric routing: Handle data packet dupes
    polypath asymmetric routing: Handle MSG1 late dupes
    polypath asymmetric routing: Handle MSG2-4 late dupes
  + site: Replace wait_timeout variable with function
  + random: Admit that we will never add error checking everywhere
  + site: Randomise key setup retry time
  + site: in "entering state RUN", say whether key is set up

 NOTES              |   7 ++-
 README             |   5 +-
 configure          |  14 ++++-
 magic.h            |   5 +-
 random.c           |  24 ++------
 secnet.h           |  29 ++++++---
 site.c             | 171 ++++++++++++++++++++++++++++++++++++++++-------------
 transform-cbcmac.c |  16 ++---
 transform-common.h |   6 +-
 transform-eax.c    |  12 ++--
 10 files changed, 194 insertions(+), 95 deletions(-)

-- 
2.11.0