GSoC 2012

[Gedare Bloom]

On Sun, Mar 25, 2012 at 3:57 PM, Julien Delange
<julien.delange at gmail.com> wrote:
> On Sun, Mar 25, 2012 at 5:35 PM, Joel Sherrill
> <joel.sherrill at oarcorp.com> wrote:
>> The main advantage is that it is small and gives you TCP/IP
>> stack capabilities in a lower footprint.
>
> I understand that point but on the other hand, can we afford to
> maintain two network stacks ? In particular, these two stacks should
> be compatible with the underlying drivers and testing/maintainance
> efforts would be more consequent with two stacks. As the libbsd
> project partly aims at providing a new stack, I am just wondering if a
> GSOC project that would focus on some part of the libbsd project would
> also make sense (with a clear definition of what should be
> integrated/done). However, please do not take it hard : if there is a
> clear reason/justification for having two separate stacks, let's go
> :-)
The only reason is application's footprint. lwIP is smaller (if less
functionality/performance). We should strive to have drivers that are
as much as possible independent of the network stack itself. That's
the real challenge in this proposed project.

-Gedare

> _______________________________________________
> rtems-devel mailing list
> rtems-devel at rtems.org
> http://www.rtems.org/mailman/listinfo/rtems-devel