Commit Graph

3746 Commits

Author SHA1 Message Date
giannozz a695a60e60 PH: reduce_io moved into control_ph
Updated control_flags module saved: CP will not compile otherwise


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3809 c92efa57-630b-4861-b058-cf58834340f0
2007-02-22 10:34:43 +00:00
giannozz ce10c8550c reduce_io => tksw; unused variable tvlocw removed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3808 c92efa57-630b-4861-b058-cf58834340f0
2007-02-22 08:45:23 +00:00
ceresoli 6e7e12382e efg removed from Makefile. (D.C.)
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3807 c92efa57-630b-4861-b058-cf58834340f0
2007-02-22 08:41:22 +00:00
giannozz 0116b4e4c4 oops, configure confusion ...
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3806 c92efa57-630b-4861-b058-cf58834340f0
2007-02-22 08:08:33 +00:00
seitsonen b0865840b8 Moved inside '../GIPAW'
apsi


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3805 c92efa57-630b-4861-b058-cf58834340f0
2007-02-22 05:14:59 +00:00
seitsonen 4683093464 One array not allocated properly
apsi


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3804 c92efa57-630b-4861-b058-cf58834340f0
2007-02-21 17:04:19 +00:00
seitsonen f70b7dc57e Changed the data types in the PAW reconstruction; now there is one major
data type for a species and an array to store this data for all species.
Seems to compile and run 'pw.x' correctly, some other problem in GIPAW/NMR...?

        apsi


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3803 c92efa57-630b-4861-b058-cf58834340f0
2007-02-21 16:54:18 +00:00
ceresoli 6854007dd6 Keep same ordering of G-vectors both at k and k+q. (D.C.)
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3802 c92efa57-630b-4861-b058-cf58834340f0
2007-02-21 14:44:05 +00:00
ceresoli 94dd1ce6d2 io_files repeated in ONLY list (D.C.)
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3801 c92efa57-630b-4861-b058-cf58834340f0
2007-02-21 14:42:24 +00:00
giannozz 0a862b7402 Added possibility (not yet activated) of keeping wavefunctions in memory.
File buffers.f90 contains replacements - with different names and syntax -
for diropn (open_buffer) and davcio (get_buffer, save_buffer). If you have
better names, please go ahead. Works for a single buffer right now.

Pitfalls: nwordwfc is now defined as the number of complex words, no longer
of real words, but only in PW/, so mixing routines from PW with routines from
other codes may run into trouble (although I didn't see any side effect).
Restart may no longer work in some specific cases.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3800 c92efa57-630b-4861-b058-cf58834340f0
2007-02-21 13:01:31 +00:00
giannozz d654a566a7 Compilation of VIB removed from Makefile. TODO updated.
configure updated following Axel's suggestion on semi-static linking


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3799 c92efa57-630b-4861-b058-cf58834340f0
2007-02-21 12:44:17 +00:00
ceresoli e2ca4e0f64 Removed the temporary hack to run with unequal number of
k-point per pool. Tested with MPICH. (D.C.)


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3798 c92efa57-630b-4861-b058-cf58834340f0
2007-02-21 10:51:12 +00:00
ceresoli 35bdbe290e Better(?) memory statistics. Unfortunately every libc fills in different
struct mallinfo fields. Tested in Linux and AIX. (D.C.)


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3797 c92efa57-630b-4861-b058-cf58834340f0
2007-02-21 09:15:19 +00:00
giannozz 467f1cf5b1 Re-added check on exit condition in a safe case
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3796 c92efa57-630b-4861-b058-cf58834340f0
2007-02-19 15:48:39 +00:00
giannozz 880ff250b7 Yet another problem with phonon + pools + lsda + wf_collect solved (maybe).
The solution is not especially satisfactory but it seems to work.

Calls to "check_stop_now" removed from c_bands. They are likely to
interfere with neb in a scf calculation; with phonons in a nscf calculation;
it is very unlikely that they will be useful in a band structure calculation.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3795 c92efa57-630b-4861-b058-cf58834340f0
2007-02-19 15:15:45 +00:00
giannozz 7059ece72a More phonon-with-pool problems
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3794 c92efa57-630b-4861-b058-cf58834340f0
2007-02-19 11:16:43 +00:00
dalcorso e59a041ceb Introduced a new input variable lkpoint_dir. If .false. it does not open
a new directory for each k point.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3793 c92efa57-630b-4861-b058-cf58834340f0
2007-02-19 09:21:42 +00:00
giannozz 73d9b28f61 Incorrect phonons with pools and lsda fixed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3792 c92efa57-630b-4861-b058-cf58834340f0
2007-02-18 16:39:14 +00:00
giannozz f4d85e0a24 Phonon dispersions were no longer working after last change to pools
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3791 c92efa57-630b-4861-b058-cf58834340f0
2007-02-18 16:37:41 +00:00
giannozz bfa88711b0 Printout of allocated memory added, warning message about non converged
eigenvalues made less dramatic


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3790 c92efa57-630b-4861-b058-cf58834340f0
2007-02-16 14:57:29 +00:00
giannozz 400cfcd37b Two meta-gga variables were not deallocated at the end of the run
(clean_pw). This was causing "already already deallocated" errors.
A third variable was not used at all and has been removed.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3789 c92efa57-630b-4861-b058-cf58834340f0
2007-02-16 13:42:48 +00:00
giannozz 17dae0696b Updated after last change (nks => nkstot)
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3788 c92efa57-630b-4861-b058-cf58834340f0
2007-02-15 16:11:07 +00:00
giannozz 1414be8bc9 Pool/k-points minor cleanup:
"nkstot" is always the total number of k-points, summed on all pools;
"nks" is defined only after pools are initialized and is the number of
k-points on the current pool. Previously there was some confusion due
to the redefinition of nks. It seems to work with no side effects, but
- it is untested for EXX
- the X phonon in Ni (example 2) does not work, but it doesn't seem
  to be related with this change

The commented lines in h_psi_meta made compilation fail on sp5.
How this was happening I don't know: it must be related to the
pre-processing of CMPLX (see include/f_defs,h). Maybe we should
get rid of it. In the meantime I removed the offending lines.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3787 c92efa57-630b-4861-b058-cf58834340f0
2007-02-15 15:05:01 +00:00
giannozz 8ff3787b51 Dependencies + minor things updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3786 c92efa57-630b-4861-b058-cf58834340f0
2007-02-15 11:53:18 +00:00
giannozz 09bf6a7e43 Makefile updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3785 c92efa57-630b-4861-b058-cf58834340f0
2007-02-15 11:51:27 +00:00
giannozz d3b89d58dd Syntax used in a few places to split lines was not accepted by all compilers.
Makefile updated


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3784 c92efa57-630b-4861-b058-cf58834340f0
2007-02-15 11:46:39 +00:00
giannozz f4ce0c52e9 Incorrect call fixed (courtesy of Javad)
Makefile and dependencies updated


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3783 c92efa57-630b-4861-b058-cf58834340f0
2007-02-14 17:02:32 +00:00
dalcorso 4083d3f0f1 Bug fix: problems with different number of k points on different pools.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3782 c92efa57-630b-4861-b058-cf58834340f0
2007-02-14 13:38:00 +00:00
degironc bf6cdc5634 elimination from CPV/metaxc.f90 of code non-specific parts that have been
moved to Modules/metagga.f90
stefano


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3781 c92efa57-630b-4861-b058-cf58834340f0
2007-02-14 08:51:32 +00:00
degironc 9c52e10ee9 missing file in the last commit. sorry.
stefano


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3780 c92efa57-630b-4861-b058-cf58834340f0
2007-02-14 08:40:23 +00:00
degironc 42a12efcc8 The Meta-GGA available on cp.x has been ported to pw.x: scf and forces,
while stress must still be done. So far, NC PP only.
SdG + Koichiro Umemoto


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3779 c92efa57-630b-4861-b058-cf58834340f0
2007-02-13 21:39:20 +00:00
giannozz d2634693b8 Minor configure updates for nec
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3778 c92efa57-630b-4861-b058-cf58834340f0
2007-02-13 15:38:19 +00:00
giannozz b9509157c4 Documentation updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3777 c92efa57-630b-4861-b058-cf58834340f0
2007-02-13 15:37:29 +00:00
dalcorso 66ed791360 *** empty log message ***
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3776 c92efa57-630b-4861-b058-cf58834340f0
2007-02-13 10:42:55 +00:00
giannozz a60c0573f4 There isn't a strong reason to try a different compiler for F77 and
F90. In at least some cases (eg F90=g95), this may create problems.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3775 c92efa57-630b-4861-b058-cf58834340f0
2007-02-09 12:42:37 +00:00
cavazzon 77f38c1e1f - bug fix in metal part
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3774 c92efa57-630b-4861-b058-cf58834340f0
2007-02-09 09:23:17 +00:00
giannozz 9757315242 gaussian broadening range used in el-ph calculation reduced by a
factor of 2: it was too big. Dependencies updated.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3773 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 16:35:56 +00:00
dalcorso 4a8073abbf Documentation update.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3772 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 16:10:24 +00:00
dalcorso c9174d2332 Added an example of phonon calculation with spin-orbit.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3771 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 14:19:14 +00:00
dalcorso 5b20f12bd4 Compatibility with the previous changes.
Please note: cgsolve_all has a new argument. It is last one and should be set
to 1 in the standard case.
star_q has two new arguments. The last two are logical and should be set
to .false. in the standard case.
If you use these two routines and I have not found your routine, please update.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3770 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 13:57:01 +00:00
dalcorso a8ba1f7888 Phonon in the noncollinear spin-orbit case. A missing routine.
(All this is very experimental).


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3769 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 13:45:34 +00:00
dalcorso 7c129ee46c Phonon in the noncollinear and spin-orbit case. Routines with more complex
changes.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3768 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 13:40:53 +00:00
dalcorso 16edbc1adc Phonon in the noncollinear and spin-orbit case. New routines.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3767 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 13:17:38 +00:00
dalcorso 69c1fd137a Phonon in the noncollinear and spin-orbit case. Routines with small changes.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3766 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 13:07:07 +00:00
dalcorso cfef6c3be5 Phonon in the noncollinear and spin-orbit case. Changes in PW.
angle1, angle2 and starting_magnetization are saved in the punch file.
The transformation of angle1 and angle2 to radiants is done in input.f90.
Clean_up of sum_band.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3765 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 12:47:41 +00:00
giannozz f4b24be5e8 I/O reduction for non-scf and electric-field calculations.
I am not especially happy with the way it is achieved but
couldn't find anything better.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3764 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 11:59:13 +00:00
giannozz e0a4d3df06 In the triclinic case (ibrav=1) the translation from traditional cell
parameters a, b, c, cosab, cosac, cosbc, to celldm was not correct.
Documentation updated. Courtesy of Marco Fornari.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3763 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 11:08:48 +00:00
giannozz f7080e7e39 example 19 updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3762 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 10:58:12 +00:00
giannozz 205056da68 Example 18 updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3761 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 10:56:45 +00:00
giannozz 3ce327cb5f example 13 updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3760 c92efa57-630b-4861-b058-cf58834340f0
2007-02-08 10:54:52 +00:00