#6211 BLOC Future : patch for support of per packet mesh ttl

[Zarro Boogs per Child]

#6211: patch for support of per packet mesh ttl
--------------------------+-------------------------------------------------
  Reporter:  mbletsas     |       Owner:  dwmw2                         
      Type:  enhancement  |      Status:  new                           
  Priority:  blocker      |   Milestone:  Future Release                
 Component:  wireless     |     Version:                                
Resolution:               |    Keywords:  libertas, ttl, mesh, Update.1?
  Verified:  0            |    Blocking:                                
 Blockedby:               |  
--------------------------+-------------------------------------------------

Comment(by jcardona):

 Replying to [comment:17 cscott]:
 > More comments on the most recent version of the patch (in addition to
 what dilinger said):
 >  * I think you need a memory fence between the write to
 mesh_sk->orig_sk_destruct and the write to sk->sk_destruct in
 do_mesh_set_mesh_ttl.  Your mesh_sks_lock doesn't protect changes to sk.

 Done.

 >  * Your definitions in linux/in.h don't match your suggested userland
 code in the header of the patch.  It looks like you need to rename both
 MESH_SO_SET_TTL and MESH_SO_GET_TTL to SO_MESH_TTL.

 Done.

 > Minor:
 >  * please remove the bits of the patch that affect pkt_len.  That might
 be a good idea, but belongs in a separate patch

 It improves readability of the code, but not the patch.  Removed.

 > , unless...
 >  * Why are you incrementing tx_packet_location and not tx_packet_length?
 Is that a bug?

 No.  tx_packet_length is the length of the 802.11 frame, excluding the
 variable length descriptor.

 > Should you be updating pkt_len

 No

 > (and is that the reason for replacing tx_packet_length with pkt_len)?

 No.  pkt_len is kept in host byte order, whereas using tx_packet_length
 requires ugly endiannes conversions.  But I agree that this should be in a
 separate patch.

 Thanks!

-- 
Ticket URL: <http://dev.laptop.org/ticket/6211#comment:21>
One Laptop Per Child <http://dev.laptop.org>
OLPC bug tracking system