scnc
/
quantum-espresso
mirror of https://gitlab.com/QEF/q-e.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
6,842 Commits 5 Branches 40 Tags 1.8 GiB
Commit Graph

6842 Commits

Author SHA1 Message Date
degironc 4d3e329b8f add ACFDT directory in the list checked for dependencies 
look for dependencies only if a directory exists


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7011 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-23 13:12:29 +00:00
degironc 95ba9a2468 printout of smearing name 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7010 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-23 13:11:10 +00:00
degironc 0c793f75c8 recent variable smearing (associated to ngauss) needs to be set when reading data from restart file. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7009 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-23 12:22:12 +00:00
dalcorso 30e0444333 Added a check that the q point mesh generated by phonon is compatible with 
the symmetry.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7008 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-22 22:35:36 +00:00
dalcorso 2c5fb306fb Introduced a new input variable to disable the mode symmetry analysis in 
the phonon.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7007 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-22 22:17:27 +00:00
dalcorso e23e877962 Avoid a possible out of bound error when the mode symmetry analysis fails 
(This should not happen).


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7006 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-22 22:10:30 +00:00
degironc 645e644ede one variables used in acfdt added in control_flags 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7005 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-20 17:27:12 +00:00
ccavazzoni 1feaa883c1 - bug fix for LSD simulation, 
WARNING: This bug was there since long time,
but only with recent update its effects become visible.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7004 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-20 15:21:29 +00:00
ccavazzoni e0a0d4014e - bug fix for LSD simulation, bug introduced with yesterday commit 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7003 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-20 15:18:43 +00:00
dalcorso 409c17d8d8 Bug fix: a wrong default in Thursday commit had very strange consequences 
in many examples.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7002 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-20 13:10:13 +00:00
nn245 ead25648b1 Deactivate probing for MKL scalapack. Automatic configuration needs to be solved. (N. Nemec) 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7001 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-20 12:13:30 +00:00
degironc ba5d657385 minor output change: smearing name (nad not the internal vaiable ngauss) 
is printed in the summary


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@7000 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-20 07:03:57 +00:00
degironc e12373c608 comment cleanup 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6999 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-20 06:59:34 +00:00
ccavazzoni 2a2094f36e - cleanup 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6998 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-19 16:06:00 +00:00
ccavazzoni f5f185e1d6 - some variables related to the OpenMP parallelization 
need to be declared also for non-OpenMP build


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6997 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-19 15:57:58 +00:00
degironc c38e0e09c0 __OPENMP precompilation fix: references to variables mytid,itid,ntids (only used with OPENMP) moved inside an #ifdef __OPENMP / #endif unit 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6996 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-19 15:49:48 +00:00
degironc b91c455bb2 clean-up of the md5_from_file routine 
R.Sabatini (via SdG)


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6995 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-19 15:30:58 +00:00
degironc fc72fa7cd8 add a (dummy) entry for acfdt.x 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6994 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-19 13:40:32 +00:00
degironc cc7fb040a7 adding some variables needed for the Adiabatic-Coupling Fluctation-Dissipation Theory part. 
They should be harmless in all other cases where are not used

SdG and nnlinh


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6993 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-19 13:24:00 +00:00
dalcorso 21b59d0f62 Introduced a new input variable: no_t_rev. 
If .true. the symmetry operations that require time reversal are not used.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6992 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-19 09:33:14 +00:00
dalcorso 032269ec73 Small change to the phonon output. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6991 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-19 08:53:58 +00:00
giannozz 2507e79d67 Documentation updated 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6990 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-18 15:14:13 +00:00
giannozz e1e66275d7 More things to do 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6989 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-18 14:40:24 +00:00
giannozz 3e16cf2b10 New routine calculating md5 seems to work properly 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6988 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-18 14:39:50 +00:00
giannozz 3a866e2304 ev.x was printing pressures in KBar and pretending them to be GPa. Now 
it prints GPa. It also prints on output the same format as on input:
a.u. if input was a.u., A if it was A.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6987 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-18 09:48:49 +00:00
giannozz 23ffbbed84 Fixed compilation problem after yesterday commmits, removed some unused 
variables


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6986 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-18 07:06:04 +00:00
ccavazzoni 3bc240b591 - OpenMP parallelization of Small Box FFT. 
time for rhov, drhov and newd now scale
  almost linerly with the number of threads.

NOTE: for the time being the spin polarized case
      has not being parallelized


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6985 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 22:16:23 +00:00
ccavazzoni f5f2072c50 - In order to reduce the fluctuations of the constant of motion 
when performing a variable cell simulation, now
  the preconditioning factors for the fourier acceleration
  are computed only once at the beginning of each run.
  The ideal would be to use the same factors from scratch, but
  this implies storing g-vectors relative to the starting cell.
  Now the constant of motion is really constant even when
  the cell is moving a part small discontinuities between
  two consecutive run.

  NOTE: it takes a lot of time to catch this problem,
  please do not reverse to the old convention of updating
  "scaled emass" at each step, even if it seems more
  appropriate.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6984 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 21:37:46 +00:00
ccavazzoni 88b95c5a0a - Fix required to make CP work even when some processor have no G vectors 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6983 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 21:18:28 +00:00
ccavazzoni 77dc7c5fd5 - An ad-hoc initialization is required for the OpenMP version of the 
fft on the small box


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6982 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 21:14:39 +00:00
ccavazzoni aa164299c9 - OpenMP parallelization of the FFT on the small box used for the core charge 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6981 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 21:12:07 +00:00
ccavazzoni 64781b146f clocks called inside a parallel region do not work properly 
(they only count the events and the time of the master thread),
 in the next future we will need a thread safe version of the clocks.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6980 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 21:08:40 +00:00
ccavazzoni 00b9d2f055 - fft wrapper subroutine updated with the OpenMP driver 
for small box.
  !!Available only for __OPENMP && __FFTW


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6979 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 21:04:54 +00:00
giannozz e4b8783505 Modified in such a way that it doesn't stop if there are additional lines 
at the end aded by OPIUM


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6978 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 20:49:15 +00:00
ccavazzoni 650037ffd0 - added two functions for OpenMP parallelization of fft on small box. 
In this case we need to split the initialization from the actual computation,
  infact the idea is to assign a whole box to a single thread and
  not parallelize the 3D fft, in other words each 3D fft on the small box
  will be performad by a single thread.

  For the same reason this parallelization can be implemented only
  with a fully scalar fft library (e.g. the fftw distributed with QE)
  otherwise the fft library (like esslsmp) trys to parallelize computation internally,
  which is not what we want.

  Preliminary results on large systems (with lots of atoms)
  shows a big boost from this kind of parallelization.

  Commit for rhov, drhov and nlcc will come shortly.

  As a general final comment, I would like to stress that for
  OpenMP parallelization of the FFT, since we usually have small
  radix (even for the larger grid), it seems not worth to
  use multithreaded fft library.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6977 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 17:17:20 +00:00
giannozz 169c41c59f Updated 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6976 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 16:24:15 +00:00
marsamos f6bd98ea17 lcoarsegrained part commented. preliminary step before erase. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6975 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 16:24:09 +00:00
marsamos c673310ef8 io part of parallelization on images moved to image_io_routines. First step 
for NEB cleaning.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6974 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 15:39:53 +00:00
obm 22842ec988 Mixup in sign convention. Fixed. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6973 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-17 09:54:16 +00:00
obm 99b6b6e4c2 Small bugfix 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6972 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-16 19:54:07 +00:00
giannozz 647fdda4d8 Re-added simple code to calculate vibrational free energy - useful for 
low-symmetry lattices not covered by QHA. Uses phonon DOS as input.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6971 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-16 14:55:47 +00:00
giannozz 774ae7016e Unused routine removed 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6970 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-16 07:35:27 +00:00
dalcorso dc851231ad Bug fix of the recent commit. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6969 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-16 07:03:23 +00:00
giannozz 8b448251e8 Minor format changes 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6968 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-15 13:33:44 +00:00
dalcorso 724267f66f Removed the constraint on the number of bands with one_atom_occupations. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6967 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-14 15:11:05 +00:00
dalcorso a12010f7d8 Prevent a possible out of bound error. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6966 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-14 15:08:17 +00:00
giannozz 69475aeeca Format for number of k-points extended to i6 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6965 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-14 10:55:34 +00:00
ccavazzoni 68701d64af - Fix in CP for units of input pressure. 
Input external pressure is in KBar and not in GPa like it was formerly in CP.
  Then from now on input value for variable "press" in cell namelist
  should be given in KBar like it is written on the documentation!


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6964 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-14 09:43:45 +00:00
degironc 2cff4d59aa more modifications related to md5 checksum 
R. Sabatini - S. de Gironcoli


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6963 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-13 10:53:51 +00:00
degironc b75413adf0 adding a few routines needed to calculate md5 checksum of a given file. 
They are intended to be used to calculate (and echo on the output) the
md5 checksum of the peudopotential files in order to be able to verify their
identity/integrity when comparing calculations performed at different
times or on different machines

R. Sabatini - S. de Gironcoli


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@6962 c92efa57-630b-4861-b058-cf58834340f0
 2010-08-13 10:50:08 +00:00