fixed a bug (uninitialized array) in pw_export.f90, some additions to the
output format (header with version number, fractionary tranmsations)
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3457 c92efa57-630b-4861-b058-cf58834340f0
probably harmless - There was some confusion with the definitions of
ZERO, ONE, RZERO, RONE, CZERO, CONE. Now: ZERO AND ONE are complex,
RZERO and RONE are real, CZERO and CONE are no more. Please check!
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3443 c92efa57-630b-4861-b058-cf58834340f0
reduce the memory usage to the very minimum. Implemented a more efficient scheme
to find all the points in each small box. The charge is rescaled to the correct
number of electrons unless the relative error is larger that 0.01% (in this case
an error message is issued suggesting to increase ecutrho). Added a sort of
Fourier filter to remove the features in the chrage density that are not
compatible with the chosen cut-off (this is used only with the small booxes).
Addded an aternative estimate of the total energy based on the
Harris-Weinert-Foulkes functional. The difference between the Kohn-Sham energy
and the HWF one provides an alternative estimate for the level of
self-consistency.
Added a few more clocks to monitor the efficiency of the small-boxes and the
time spent to calculate the Hatree potential and the Exchange-Correlation
potential (the latter can be pretty high).
C.S.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3437 c92efa57-630b-4861-b058-cf58834340f0
lapack_mkl.f does not seem to be needed any longer starting from mkl v6 or so
Since earlier versions are not supported by configure anyway I removed it
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3436 c92efa57-630b-4861-b058-cf58834340f0
blas and lapack libraries are specified solely in BLAS_LIB and LAPACK_LIB,
no more in MYLIB. A Lapack library is always compiled but used only if
needed. Problem with conflicting essl/lapack on cineca sp5 solved (i hope)
Re-run configure before complaining that nothing works !
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3435 c92efa57-630b-4861-b058-cf58834340f0
The rectangular base (a,b) is centered while the monoclinic angle<>90 is
between b and c. Problem pointed out by Young-Su Lee @ mit .
SdG
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3430 c92efa57-630b-4861-b058-cf58834340f0
libcp.a respectively), used to load the executables. This removes
the need to specify all needed objects in many different makefiles.
Works with 95, ifort, xlf. Please let me know if there are problems
with more exotic or lousy compilers.
A few minor things fixed as well.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3428 c92efa57-630b-4861-b058-cf58834340f0
the code symmetrizes the atomic (and iif needed the cell) configuration
before stopping, so as to allow to restart from a symmetric configuration
if desired.
SdG
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3425 c92efa57-630b-4861-b058-cf58834340f0
The include file for C routines is generated by configure,
in a way that should be very general and robust. Beware:
you need to re-run configure to regenerate include/c_defs.h
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3423 c92efa57-630b-4861-b058-cf58834340f0
I added some comments here and there and slightly modified the output,
removing the 'band sum' which seems to me of no interest to anybody,
leaving the various contributions whose sum adds up to the total energy.
The nfamous 'correction for metal' is labeled as ' "-TS" contribution ',
so making clear that there can be a relation with the free energy
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3422 c92efa57-630b-4861-b058-cf58834340f0
even when a and b should be EXACTLY equal. Some compilers manage to
make them slightly differnt anyway. Courtesy of E.A. Menendez.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3420 c92efa57-630b-4861-b058-cf58834340f0
F77_FUNC and F77_FUNC_ preprocessor macro . It is easy now to select the
good functions with "configure" (it is still done with -D flags though)
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3416 c92efa57-630b-4861-b058-cf58834340f0
It had gone unnoticed for a long time, but the recent changes
made its presence noticeable via a rather obscure failure.
C_POINTER is defined integer*N instead of integer(kind=N).
The former definition is obsolescent but always valid, while the latter
may fail for compilers (ex. NAG) not using the length in bytes as kind
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3413 c92efa57-630b-4861-b058-cf58834340f0
make.depend are now in the CVS and shouldn't be updated unless
needed. Configure no longer runs makedeps.sh. Type 'make depend'
to regenerate dependencies.
Remove all */make.depend files before updating!
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3409 c92efa57-630b-4861-b058-cf58834340f0
dalcorso