[Olsr-dev] Bug Secure Plugin - Endianness ?

Rusty Haddock (spam-protected)
Sat Jul 2 00:07:50 CEST 2011


On Sat, Jun 25, 2011 at 6:35 AM, Henning <(spam-protected)> wrote:

> On Saturday 25 June 2011 06:58:46 Rusty Haddock wrote:
> > Hi!
> >
> >     I'm new to this mailing list but I've been one of developers with
> > the Ham Radio HSMM-MESH project down in Austin, TX for two yearas now.
> >...
>
>     The code has been running on my personal mesh of 6 nodes for about 3
> weeks.
>  > Most of my mesh is composed of WRT54G's although it has been running on
> one or
> > two little Atheros-based nodes as well as my x86 laptop.
>
> So your fix does run both for little and bit endian devices ? (Maybe even
> for
> networks mixed with both) ? :)
>
> Henning
>

Sorry for the delay as I've been busy with a lot of other things this week.

Yes, my fixes runs on both little and big endian devices allowing mixed
networks.
There's no need to #ifdef things depending on which devices you're compiling
for.
The changes essentially affect the little endian devices such as the
WRT54G's
with the Broadcom MIPSEL.  The compiler will generate more code to flip the
bytes around to network order for the devices.

On the big endian devices, such as the Atheros MIPS, the changes compile
down
to nothing since the hton[ls]() and ntoh[ls]() functions/macros are
basically null on
the big endian devices since they natively use the network byte ordering.

For those really weird machines with weird byte ordering (are there any?)
these
changes should let them play nicely with the other kids now, at least if
they're using
the secure plugin.

I've just pushed the changes so have a look at my changes and see what you
think.

Thanks!

     -Rusty-
-- 
   _____                Rusty Haddock  <=>  AE5AE
|\/   o \   o             Way out yonder in the
|   (  -<  O o      Van Alstyne (TX) Metropolitan Area
|/\__V__/       Math illiteracy affects 7 out of 5 people!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.olsr.org/pipermail/olsr-dev/attachments/20110701/effe6f60/attachment.html>


More information about the Olsr-dev mailing list