Hi
I want to add some concerns to this issue. I have tried that and I can't
get Amer to compile either. I have the same problem as Ross. I have tried
both intel and gnu and I get the same problem. I am compiling the version
from the git tree. I did a git clone just now before configuring and
compiling before writing this email.
These are the specifications for the system I am using:
Red Hat 5.8
the intel compiler 10.1
gnu compilers 4.1.2
mpich2 1.2.1p1
Example of errors with intel
ifort -c -DBINTRAJ \
\
-ip -O3 -axSTPW -FR
-I/server-home2/romelia/amber-mine/include
-I/server-home2/romelia/amber-mine/include \
-o rism3d_fft.o rism3d_fft.F90
fortcom: Error: rism3d_fft.F90, line 319: A pointer dummy argument with the
INTENT(IN) attribute shall not appear as an actual argument if the
associated dummy
argument has the INTENT(OUT) or INTENT(INOUT) attribute. [INR]
inr(:,1),this%grid%ngk(3:1:-1),1,product(this%grid%ngk),&
---------^
Example of errors with gnu
cd ../rism && \
make amber_rism_interface.NAB.o && \
cp -f amber_rism_interface.NAB.o ../sff/. ; \
make[3]: Entering directory
`/server-home2/romelia/amber-compiled/AmberTools/src/rism'
gfortran -c -DBINTRAJ \
\
-O3 -ffree-form
-I/server-home2/romelia/amber-compiled/include
-I/server-home2/romelia/amber-compiled/include \
-o rism_report_c.o rism_report_c.F90
gfortran -c -DBINTRAJ \
\
-O0 -ffree-form
-I/server-home2/romelia/amber-compiled/include
-I/server-home2/romelia/amber-compiled/include \
-o fftw3.o fftw3.F90
Error: Can't open included file 'fftw3.f03'
In file fftw3.F90:33
use, intrinsic :: iso_c_binding
1
Error: Unclassifiable statement at (1)
>
> On 2012-03-16, at 8:57 PM, Ross Walker wrote:
>
>> Hi Tyler,
>>
>>> I had explicitly tested this before it seems something has changed.
>>>
>>> To be clear, are you building from the Git repo or from the tar ball?
>>
>> This is from the git repo - (git clone
>> gitosis.git.ambermd.org:amber.git) as
>> of this morning.
>>
>
> Will the next version of the release be taken from the Git repo or has the
> released version already forked?
>
>>> configure should have died with an error message and explicit
>>> instructions on how to configure without RISM. From your message it
>>> sounds like this is not happening. Can you confirm that there is no
>>> error message?
>>
>> Configure doesn't die anymore. It prints the following message:
>>
>> ./configure gnu
>>
>> ....
>> Obtaining the gnu suite version:
>> gcc -v
>> The version is 4.1.2
>> Skipping the parts of AmberTools that depend on fftw3
>> ....
>>
>> and sets $RISM = no
>>
>
> I think we should have a consistent policy as to what happens when some
> functionality is not available with a particular compiler or whatever.
> IMO, I think configure should _always_ die with an error message and
> explicit instructions on how to proceed. My previous error message would
> return the user's original configure command modified such that RISM would
> not be built by adding '-norism' and removing any instances of '-rism'.
> This way they could copy and paste to quickly get to the next step but
> also knew exactly what functionality was missing.
>
> I will get this working with the current warning message instead of an
> error message for now. I'll try to get it done tonight but it may be as
> late as tomorrow morning.
>
> Tyler
>
>
>
>
>
>
> _______________________________________________
> AMBER-Developers mailing list
> AMBER-Developers.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber-developers
>
--
****************************************
Romelia Salomon
Walker Group
398 San Diego Supercomputing Center
UC San Diego
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Mar 16 2012 - 19:00:03 PDT
