WRF 4.1 Warnings and Errors with Intel compiler

[JCollins]

Hi all.

I am trying to compile WRF 4.1 with the Intel compiler with version 18.0.5 20180823. The compilation finishes successfully, but when reviewing the output file I find several warnings and several errors.

The warnigs are repeated a lot of these two types:

gen_comms.c(1229): warning #303: explicit type is missing ("int" assumed)

REGISTRY WARNING: o3rad: o3rad_b is not a variable or number; ignoring it

And the errors go above all like this:

icc: error #10236: File not found:  'all'
icc: error #10236: File not found:  'big_endian'
make[2]: [bobrand.o] Error 1 (no tiene efecto)

For the compilation I used "Suggestions for timing improvements from Craig Mattocks" that in previous compilations worked very well.

I've done a couple of tests and I'm having some performance problem, I do not know if warnings and errors can influence or can be eliminated with a flag. Attachment "configure.wrf" and "compile.log".

Thank you.

 

[Ming Chen]

Would you please compile WRF with just one processor (e.g., setenv J "-j 1" in csh)? In this case we will know when and where the first error popped up. Other errors may derive from the first error.

 

[JCollins]

I have recompiled the model with a single processor. I attach the new file.

 

[JCollins]

Does anyone know how to solve the error?

 

[davegill]

John,
When I tailed your compile.log, here is what I see:

==========================================================================
build started:   mié may  8 18:03:13 WEST 2019
build completed: mié may 8 18:54:44 WEST 2019

--->                  Executables successfully built                  <---

-rwxr-xr-x. 1 root root 52843320 may  8 18:54 main/ndown.exe
-rwxr-xr-x. 1 root root 52890968 may  8 18:54 main/real.exe
-rwxr-xr-x. 1 root root 52168912 may  8 18:54 main/tc.exe
-rwxr-xr-x. 1 root root 59592944 may  8 18:53 main/wrf.exe

==========================================================================

It looks like the build is successful.
Dave

 

[JCollins]

Correct, the compilation ends correctly. But then my question is, to what can these errors be due? Do they influence the final behavior of WRF?