Google Summer of Code projects

[Scott Adams]

Hi;

I have only just this week woken up to the possibility of doing a 
project under the GSoC program.  Though I do not have any previous 
experience with RTEMS specifically, I have over 25 years of mostly 
embedded development experience - from hardware design through 
high-level architecture of multi-processor systems (I'm not your typical 
student).

Anyway, I am trying to complete an application for one or possibly two 
different RTEMS projects, and hope to submit it/them tomorrow.

This is primarily a heads-up (two postings requesting that applications 
be submitted NOW); I am not even close to ready to submit - sorry.

The projects I'm looking at are: the BSD TCP/IP stack update (I have 
over 10 years of TCP/IP protocol implementation and porting experience), 
lwip and BSD TCP/IP driver rationalization, TinyRTEMS, Ubuntu build 
tools support, Eclipse build/debug integration, and ArgoUML modeling/C++ 
code generation.

Here are my thoughts:

    * I am a Ubuntu user/fan, and intend to do any project under Ubuntu,
      so Ubuntu platform support would come for free with any project I
      do (I have discovered some issues with the present instructions
      for installing the RTEMS tools packages under Ubuntu - I'll post
      the specifics tomorrow).  General debian support may be trivial
      from this.
    * I am also an Eclipse CDT (Ganymede) user for embedded development
      (recently using FreeRTOS on Atmel AT91SAM7 SoC parts).  Some
      refinement/development of that platform would also come included
      in whatever project I do.
    * I am a fan of UML, and suggest that BOUML (UML 2, supporting XMI
      2.1 as required for Eclipse EMF) be considered for a UML/modeling
      design tool in place of ArgoUML, which is UML 1.x only.  BOUML
      generates C++ (along with IDL and Java, plus others), and can be
      nicely tailored for RTEMS.  I don't know how much of this I could
      work into any project (specifically the BSD TCP/IP stack and/or
      lwip driver), but I would certainly try.
    * The other project of interest to me (besides the BSD TCP/IP stack
      update and/or lwip -TCP/IP driver rationalization) is the
      TinyRTEMS project.  I have a great deal of experience with small
      Real-Time kernels in this range (FreeRTOS is one example), and
      like the goals.  The same free stuff (Ubuntu, Eclipse, possible
      BOUML integration) applies.

Anyway, I'm interested in any feedback before I submit my application(s) 
tomorrow.

Scott