[Insight-developers] Wrapping craps out on PowerPC Macintosh/OSX?

[kent williams]

Always good to reply to your own posts ...
Apparently the first error makes an incorrect assumption that there's a
missing prototype for sigalstack, that isn't missing, and furthermore has a
different signature than the one in the PPC 10.5.3 headers.

The second set of errors looks like it's asking for Intel register names on
a PPC machine.

I've gone this long without learning any more than I had to about wrapping,
someone rescue me from having to understand the internals of how Cableswig
gets built!

If it matters, this is within a Slicer3 build...


On 6/5/08 10:09 AM, "kent williams" <norman-k-williams at uiowa.edu> wrote:

> I can patch around this, and I know it's probably a rare system to build on,
> but ... PowerMac OS X 10.5.3
> 
> What's worse, it's in the hacked up GCC used for CableSwig.
> 
> Building C object
> Utilities/CableSwig/GCC/gcc/CMakeFiles/backend.dir/tree-inline.o
> Insight/Utilities/CableSwig/GCC/gcc/config/rs6000/host-darwin.c:39: error:
> conflicting types for 'sigaltstack'
> /usr/include/signal.h:89: error: previous declaration of 'sigaltstack' was
> here
> Insight/Utilities/CableSwig/GCC/gcc/config/rs6000/host-darwin.c: In function
> 'segv_handler':
> Insight/Utilities/CableSwig/GCC/gcc/config/rs6000/host-darwin.c:81: error:
> 'struct __darwin_mcontext' has no member named 'ss'
> Insight/Utilities/CableSwig/GCC/gcc/config/rs6000/host-darwin.c:130: error:
> 'struct __darwin_mcontext' has no member named 'es'
> Insight/Utilities/CableSwig/GCC/gcc/config/rs6000/host-darwin.c:131: error:
> 'struct __darwin_mcontext' has no member named 'ss'
> make[2]: *** 
> [Utilities/CableSwig/GCC/gcc/CMakeFiles/backend.dir/config/rs6000/host-darwin.
> o] Error 1
> make[2]: *** Waiting for unfinished jobs....
> make[1]: *** [Utilities/CableSwig/GCC/gcc/CMakeFiles/backend.dir/all] Error 2