g95 netcdf / mac architectures

[Ryan Schmidt]

Ok. Don't forget to reply all so your reply goes to the mailing list too, not just to me.


On Mar 22, 2010, at 06:57, Felipe Pimenta wrote:

> Hi Ryan,
> 
> Yes, just warnings. Yes, the build completes, and the executable runs
> well, in both machines. I wrote to G95 team about it, but I have not
> received their answer yet. I will let you know if they reply to this
> problem.
> 
> Thanks,
> 
> F.
> 
> On Mon, Mar 22, 2010 at 7:19 AM, Ryan Schmidt wrote:
>> 
> 
>> On Mar 21, 2010, at 12:49, Felipe Pimenta wrote:
>> 
>>> The PowerPC compilation is smooth. The Intel generates a long list of
>>> warnings during the compilation. Here's a piece of these warnings:
>>> 
>>> g95    -O5   -c -o areas_masks.o areas_masks.f
>>> areas_masks.f: In function ‘areas_masks__’:
>>> areas_masks.f:44: warning: will never be executed
>>> areas_masks.f:37: warning: will never be executed
>>> areas_masks.f:32: warning: will never be executed
>>> areas_masks.f:25: warning: will never be executed
>>> areas_masks.f:17: warning: will never be executed
>>> g95    -O5   -c -o advave.o advave.f
>>> advave.f: In function ‘advave_’:
>>> advave.f:160: warning: will never be executed
>>> advave.f:151: warning: will never be executed
>>> ...etc
>>> 
>>> I've noticed that some lines are in the start of a "do loop". Both
>>> generate an executable, however.
>>> 
>>> Are the warnings related to their architectures? Any hints on how to fix this?
>> 
>> Warnings only -- not errors? The build completes?
>> 
>> You may want to report this to the developers of the software.