[R-SIG-Mac] gfortran

[Paul Roebuck]

On Fri, 27 Jan 2006, Simon Urbanek wrote:

> On Jan 27, 2006, at 12:31 PM, Paul Roebuck wrote:
>
>> The gfortran referenced on the TigerGCC wiki page is 4.0.0(5202)
>> for Xcode 2.1. I loaded this on my system before remembering I'm
>> running Xcode 2.2.1. My gcc version is 4.0.1(5250) which is newer
>> than the full replacement tarball 4.0.1(5224) as well.
>>
>> Are these compatible or is there a later (working) gfortran that
>> goes with the build on my iMacG5?
>
> The problem is that the answer to the latter is "no". Apple's branch
> has broken gfortran a while ago. So right now there are basically two
> alternatives - use old Apple-branch gfortran or use the FSF build (I
> tried to port FSF changes to Apple's build but failed - anyone who
> succeeded, please let me know).

I continue to hear Apple's branch has broken gfortran, though never
hear specifically what is broken about it. I don't mind filing a
bug report with them too (always heard their Java engineers fixed
what got the most bug reports first). I imagine there are enough
people reading this list to file enough reports to at least get
someone to look at it. Or possibly get someone with expiring
"incidents" to use one on this. Do we have enough information
to say specifically what's wrong with gfortran?

----------------------------------------------------------
SIGSIG -- signature too long (core dumped)