[Bioc-devel] gfortran instead of g77 ?

[Herve Pages]

Hi Korbinian,

(I've cc'ed to bioc-devel where, I think, we can benefit from more 
feedback.)

Korbinian Strimmer wrote:

>Dear Herve,
>
>  
>
>>  o The -Wall flag is back. We've been using it for the 1.8 builds
>>    since February.
>>    It's back in the 1.9 builds too. I will also try to add the
>>    -pedantic flag if this can lead to even more usefull compiler
>>    messages.
>>    
>>
>
>I have another suggestion for BioC 1.9 and 1.10:  would it be possible
>that in the Linux tool chain instead of the (obsolete) "g77" you
>now use the "gfortran" compiler?
>
>"gfortran" is the official Fortran 95 compiler and is part of
>gcc 4 (http://gcc.gnu.org/wiki/GFortran).
>
>R itself compiles perfectly with "gfortran", but unfortunately not all
>BioC packages (e.g. impute). So having "gfortran" as default this would
>probably lead to more compatible Fortran code...
>  
>

When we build R on our Build System machines, we let the 'configure' script
pick up the right compilers for us.
The default search for the Fortran compiler is described here
  http://cran.r-project.org/doc/manuals/R-admin.html#Using-FORTRAN)

So, g77 will be picked up on a Linux systems with gcc 3 (e.g. SUSE 9.2) and
it will be gfortran on a Linux system with gcc 4 (e.g. SUSE 10.0).

Compare the BUILD output for the DNAcopy package on gopher5 (SUSE 9.2)
vs gladstone (SUSE 10.0):
  http://bioconductor.org/checkResults/1.9/gopher5/DNAcopy.buildsrc.html
  http://bioconductor.org/checkResults/1.9/gladstone/DNAcopy.buildsrc.html

To make this more clear, more details about the platforms used on our
Build System machines will be available on the build/check report soon...
Thanks for your feedback.
Best,

H.

-- 
------------------------
Hervé Pagès
E-mail: hpages at fhcrc.org
 Phone: (206) 667-5791
   Fax: (206) 667-1319