> That compiler is the Apple one... switching away from it is essentially 
> impossible (I don't know of anyone, anywhere that has ever been able to get 
> Fortrran and C++ to link together on OSX using non Apple GCC).  The Intel 
> compilers used to be an option... but they have their own issues (like not 
> updating them properly for new versions of XCode, etc.).

Just curious. I have been using the compilers from hpc.sourceforge.net
with open-mpi+petsc+slepc+arpack+libmesh without any problems in
compile/link/run cycle on my macbook. But I do not use Xcode that much
though and it still uses the in-built apple version of gcc. Have you
tried using the hpc compilers in your macs ? The gcc version is at 4.6
on those.

Vijay

On Thu, Jun 16, 2011 at 1:00 PM, Derek Gaston <[email protected]> wrote:
>
> On Jun 16, 2011, at 11:46 AM, Roy Stogner wrote:
>
>> I'm just saying that, even if we do make the switch, you should
>> upgrade your compiler too.  If it flubbed std::fill that badly then
>> there's probably lots of other stuff it's not optimizing as well as it
>> should be.
>
> You have to remember that we use Macs....
>
> That compiler is the Apple one... switching away from it is essentially 
> impossible (I don't know of anyone, anywhere that has ever been able to get 
> Fortrran and C++ to link together on OSX using non Apple GCC).  The Intel 
> compilers used to be an option... but they have their own issues (like not 
> updating them properly for new versions of XCode, etc.).
>
> I'm hoping that with Lion they will give us either updated GCC, or a 
> CLANG/LLVM combo that works.  Or maybe someone will finally release a third 
> party compiled version of GCC that can link Fortran and C++ together... I 
> guess I can always dream....
>
>> It'd be nice to do a change like this ATLAS-style, where we test
>> performance at configure-time and pick what to use... not sure if
>> that's safe for casual use, though; I'd hate to get a grossly
>> suboptimal option just because I happened to be running another
>> process that interfered with timing while the optimal implementation
>> was being tested.
>
> Interesting idea, but I agree with your conclusion.
>
> We could do some ifdefs to catch GCC 4.2... but I'm not sure it's necessary.  
> Like you point out, memset doesn't seem to ever produce _terrible_ 
> timings.... just less good in some cases....
>
> Derek
> ------------------------------------------------------------------------------
> EditLive Enterprise is the world's most technically advanced content
> authoring tool. Experience the power of Track Changes, Inline Image
> Editing and ensure content is compliant with Accessibility Checking.
> http://p.sf.net/sfu/ephox-dev2dev
> _______________________________________________
> Libmesh-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/libmesh-devel
>

------------------------------------------------------------------------------
EditLive Enterprise is the world's most technically advanced content
authoring tool. Experience the power of Track Changes, Inline Image
Editing and ensure content is compliant with Accessibility Checking.
http://p.sf.net/sfu/ephox-dev2dev
_______________________________________________
Libmesh-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-devel

Reply via email to