Yambo Community Forum

Hello Myrta,

Unfortunately, this did not work for me. I am still having the same problem in which the compile breaks at the slatec Makefile. I have tried both the -ffreestanding and the -P arguments.

Thank you very much for your help,

Ryan

Hallo Ryan

can you upload the config.log. I can try to give a look if I see some possible solution, or recognise what causes the problem

Certainly.

Thank you,

Hallo Ryan

What I posted was related to a slightly different problem indeed. Sorry for misleading you.

From a close inspection of the log and comparing what I got from my PC which has the same gcc version
What causes problems is
C_AS_CPP_FLAGS='-I/opt/local/openmpi/1.8.4/gcc/4.8.3//include'
where do you specify this flag?
In my case I do not specify that flag and from the configure it takes the value
C_AS_CPP_FLAGS='-P'

in the config/report then I got
[CPP] gcc -E -P

Try to leave the C_AS_CPP_FLAGS unspecified so that the configure can choose based on the compiler you have on your system

Best
m

Good evening Myrta,

Unfortunately this did not resolve the problem. C_AS_CPP_FLAGS was initially being set when I loaded the openmpi module and I unset it such that configure did set C_AS_CPP_FLAGS='-P'. But the same problem was encountered when it went to build slatec.

I have attached the config.log from this attempt. I am also attaching the script that I am using to set up the build environment in case that might be of some use.

Thanks again for all your help,

Ryan

Hallo Ryan

I cannot spot anything else that can cause the problem.
One thing you can test further is if one of the other options (double precision or openmp) interfering for some reason. Another possibility is to unset all environment variables except maybe CC=gcc and FC=gfortran and see if this compiles. Of course it may give a code missing feature you want, but if it can help to understand what is wrong.

Best
m

NB As you are using openmpi you can use the --enable-openmpi for OpenMPI specific code instructions