[R-SIG-Mac] native binary for macintel machine

[Byron Ellis]

Its all good, I'm using your 2.3 nightly builds since it appears  
someone did something to useDynLib. I'm sort of hoping I can use it  
to resolve symbols more easily.... I need to try libffi again to see  
if I can get the objc/ffi stuff working on x86, the Darwin/x86 code  
wasn't working last time I checked.

On Jan 3, 2006, at 2:52 PM, Simon Urbanek wrote:

> On Jan 3, 2006, at 4:21 PM, Jan de Leeuw wrote:
>
>> I have NOT been able to compile a native gfortran from the apple  
>> cvs (low priority for apple) and i have NOT been able to build R  
>> using the intel C and fortran compilers (mostly configure problems).
>
> I was struggling with gfortran on the IntelMac for a while, too.  
> The only (and I tried many, including backporting stuff, trust me)  
> configuration using gcc that (sort of) works is FSF build from the  
> 4.0.x branch.
>
> Jan, can we compare some benchmarks? I suppose the Intel compiler  
> should be much faster ... (and no, I still don't have a good  
> benchmark example - we can use yours).
>
> I'm working on R 2.2.1 release builds for ppc64 and i386 right now  
> (batteries included ;)), so stay tuned. [Yes, ppc release will be  
> out very soon, too - just the GUI localizations are a tad behind  
> schedule - my apologies - you can get the framework from the  
> nightlies page for now].
>
> Cheers,
> Simon
>

---
Byron Ellis (ellis at stat.harvard.edu)
"Oook" -- The Librarian