[Boinc_opt] Compiling boinc_public / GCC 4.0 / Darwin 8.1.0

[rick]

Aha! Someone in the exact same situation as me.

This post to a GNUstep mailing list:
     http://lists.gnu.org/archive/html/discuss-gnustep/2003-11/ 
msg00281.html
indicates that you should be passing "-framework CoreFoundation"  
somewhere.

I'm not familiar with the GNU build system so I'm currently trawling  
through the autoconf / automake manuals. Probably at least another  
week before I get any results.

I'm working on converting seti_analyze and related functions to use  
vDSP and add some of my own Altivec enhanced routines (e.g.  
v_ChirpData). I've been grappling with the build system so my code is  
preliminary and probably won't compile. If you're interested, I'll  
put it on some webspace.



r i c k

On 17 May 2005, at 16:23, Shiloh Heurich wrote:

> ...
>
> At this point, the client builds; however running it gives the  
> following output :
>
> luh:/usr/local/build/boinc_public shiloh$ ./client/boinc --version
> dyld: Symbol not found: ___CFConstantStringClassReference
>   Referenced from: /usr/local/build/boinc_public/./client/boinc
>   Expected in: flat namespace
>
> Trace/BPT trap
>
> ###############
>
> So, my first guess is that it is somehow related to the dynamic  
> linker under Darwin. Any thoughts?
>
> Regards,
> Shiloh Heurich
>
>
>