> Looks like a nice dev board they have made for that competition.
> Lets hope that the mainline GCC supports Coretex M3 soon, and after that
> WinARM, GNU ARM and YAGARTO too.
It seems that the "mainline/trunk" gcc source-code in the CVS does not
yet include options/code for the ARMv7-architecture (like i.e. Cortex M3
CPUs). As far is I can see support is currently just available in the
"CSL"-branch maintained by Codesourcery which is the codebase for the
Codesourcery "arm-none-eabi"-distribution (of cause) and seems to be the
codebase for the gcc-binaries in Rowley's Crossworks too (not sure - I
don't have Crossworks).
Bintuils 2.17 (and the binutils created from a CVS-checkout as in WinARM
6/06) seem to support ARMv7/Cortex M3 already.
I have found some pathes to add support in the gcc-Mailing-List provided
by someone working for Codesourcery but like Fordp I hope the next
official gcc-relase will include the support since I try to avoid
patching the sources if possible. Esp. since the Codesourcery
distribution can "co-exist" with WinARM (and maybe GNUARM or Yagarto).
Maybe the "mainline" gcc 4.1.1 (as in WinARM) can support Cortex M3
already somehow when using the ARMv6-architecture option together with
-mthumb. But I'm still not sure about the stack-aligment needed for
proper exception-handling. Feedback from someone who could test this on
hardware or with a ARMv7-simulator is welcome.