svn version of bader compilling failed.

[hongyi]

Hi all,

I use the following command to obtain the svn version of bader:

svn co http://theory.cm.utexas.edu/svn/bader

And then compile it with the following command and meet errors as follow:

$ make -f makefile.lnx_ifort
ifort -O2 -check bounds -c kind_mod.f90
ifort -O2 -check bounds -c matrix_mod.f90
ifort -O2 -check bounds -c ions_mod.f90
ifort -O2 -check bounds -c options_mod.f90
ifort -O2 -check bounds -c charge_mod.f90
ifort -O2 -check bounds -c chgcar_mod.f90
chgcar_mod.f90(89): remark #8290: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+3'.
READ(100,'(3(2X,1F8.6))') ions%r_dir(i,:)
------------------------^
chgcar_mod.f90(175): remark #8290: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+3'.
WRITE(100,'(3(2X,1F8.6))') (ions%r_dir(i,:) , i=1,ions%nions)
-----------------------^
chgcar_mod.f90(178): remark #8291: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+7'.
WRITE(100,'(5(1X,E17.11))') (((chg%rho(n1,n2,n3), &
----------------------^
ifort -O2 -check bounds -c cube_mod.f90
ifort -O2 -check bounds -c io_mod.f90
ifort -O2 -check bounds -c bader_mod.f90
ifort -O2 -check bounds -c voronoi_mod.f90
ifort -O2 -check bounds -c critpoint_mod.f90
ifort -O2 -check bounds -c multipole_mod.f90
ifort -O2 -check bounds -c weight_mod.f90
ifort -O2 -check bounds -c main.f90
rm -f bader
ifort -static main.o -o bader kind_mod.o matrix_mod.o ions_mod.o options_mod.o charge_mod.o chgcar_mod.o cube_mod.o io_mod.o bader_mod.o voronoi_mod.o critpoint_mod.o multipole_mod.o weight_mod.o
main.o: In function `MAIN__':
main.f90:(.text+0xdc): undefined reference to `brute_force_'
make: *** [bader] Error 1

Any hints?

Regards

[wc5879]

There was a non-functional version at one point in history.
If you get the latest version everything should be fine.


[quote="hongyi"]Hi all,

I use the following command to obtain the svn version of bader:

svn co http://theory.cm.utexas.edu/svn/bader

And then compile it with the following command and meet errors as follow:

$ make -f makefile.lnx_ifort
ifort -O2 -check bounds -c kind_mod.f90
ifort -O2 -check bounds -c matrix_mod.f90
ifort -O2 -check bounds -c ions_mod.f90
ifort -O2 -check bounds -c options_mod.f90
ifort -O2 -check bounds -c charge_mod.f90
ifort -O2 -check bounds -c chgcar_mod.f90
chgcar_mod.f90(89): remark #8290: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+3'.
READ(100,'(3(2X,1F8.6))') ions%r_dir(i,:)
------------------------^
chgcar_mod.f90(175): remark #8290: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+3'.
WRITE(100,'(3(2X,1F8.6))') (ions%r_dir(i,:) , i=1,ions%nions)
-----------------------^
chgcar_mod.f90(178): remark #8291: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+7'.
WRITE(100,'(5(1X,E17.11))') (((chg%rho(n1,n2,n3), &
----------------------^
ifort -O2 -check bounds -c cube_mod.f90
ifort -O2 -check bounds -c io_mod.f90
ifort -O2 -check bounds -c bader_mod.f90
ifort -O2 -check bounds -c voronoi_mod.f90
ifort -O2 -check bounds -c critpoint_mod.f90
ifort -O2 -check bounds -c multipole_mod.f90
ifort -O2 -check bounds -c weight_mod.f90
ifort -O2 -check bounds -c main.f90
rm -f bader
ifort -static main.o -o bader kind_mod.o matrix_mod.o ions_mod.o options_mod.o charge_mod.o chgcar_mod.o cube_mod.o io_mod.o bader_mod.o voronoi_mod.o critpoint_mod.o multipole_mod.o weight_mod.o
main.o: In function `MAIN__':
main.f90:(.text+0xdc): undefined reference to `brute_force_'
make: *** [bader] Error 1

Any hints?

Regards[/quote]

[hongyi]

[1] By saying the non-functional latest version, do you mean the following version or not?

Version 0.28a 07/12/12

[2] Still, I cann't understand why the svn version cann't be compiled correctly? Considering that the svn version is the latest version, why it has unfixed issue till now?

Regards

[wc5879]

[1] 0.28a means a version that we give to the software, but svn versions are different, and 0.28a is not a svn version. There may be many svn versions for 0.28a, as svn versions are saves along the path of development.

[2] I'm not sure what you mean ""unfixed issue till now". I was adding new functions, and the version that I committed to svn repo cannot compile because of an incomplete clean-up of unnecessary codes. Can the program compile now?

[quote="hongyi"][1] By saying the non-functional latest version, do you mean the following version or not?

Version 0.28a 07/12/12

[2] Still, I cann't understand why the svn version cann't be compiled correctly? Considering that the svn version is the latest version, why it has unfixed issue till now?

Regards[/quote]

[hongyi]

Thanks a lot, it can compile succesfully now.

[hongyi]

See the following compilling log:


werner@debian:~/vasp-refind/HenkelmanGroup/svn/bader$ make -f makefile.lnx_ifort
ifort -O2 -check bounds -c kind_mod.f90
ifort -O2 -check bounds -c matrix_mod.f90
ifort -O2 -check bounds -c ions_mod.f90
ifort -O2 -check bounds -c options_mod.f90
ifort -O2 -check bounds -c charge_mod.f90
ifort -O2 -check bounds -c chgcar_mod.f90
chgcar_mod.f90(89): remark #8290: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+3'.
READ(100,'(3(2X,1F8.6))') ions%r_dir(i,:)
------------------------^
chgcar_mod.f90(175): remark #8290: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+3'.
WRITE(100,'(3(2X,1F8.6))') (ions%r_dir(i,:) , i=1,ions%nions)
-----------------------^
chgcar_mod.f90(178): remark #8291: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+7'.
WRITE(100,'(5(1X,E17.11))') (((chg%rho(n1,n2,n3), &
----------------------^
ifort -O2 -check bounds -c cube_mod.f90
ifort -O2 -check bounds -c io_mod.f90
ifort -O2 -check bounds -c bader_mod.f90
ifort -O2 -check bounds -c voronoi_mod.f90
ifort -O2 -check bounds -c critpoint_mod.f90
ifort -O2 -check bounds -c multipole_mod.f90
ifort -O2 -check bounds -c weight_mod.f90
ifort -O2 -check bounds -c main.f90
rm -f bader
ifort -static main.o -o bader kind_mod.o matrix_mod.o ions_mod.o options_mod.o charge_mod.o chgcar_mod.o cube_mod.o io_mod.o bader_mod.o voronoi_mod.o critpoint_mod.o multipole_mod.o weight_mod.o


As you can see, there are three lines, i.e., the lines including ``remark #'', does this harm?

Regards

[graeme]

The CHGCAR format is set by the vasp folks and not by us.

[hongyi]

Got it. Thanks a lot.

Regards