Commit Graph

5320 Commits

Author SHA1 Message Date
dalcorso 7db7c27f61 Improved output messages.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5413 c92efa57-630b-4861-b058-cf58834340f0
2009-02-14 09:05:01 +00:00
dalcorso 097596216d Phonon cleanup. The preconditioning variable eprec is calculated in a single
place. h_diag is calculated outside the loop on the perturbations.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5412 c92efa57-630b-4861-b058-cf58834340f0
2009-02-14 07:46:55 +00:00
dalcorso 94f4212854 Minor phonon cleanup. The rotation of dbecsum_nc is done in a single routine.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5411 c92efa57-630b-4861-b058-cf58834340f0
2009-02-14 06:43:40 +00:00
dalcorso 01dc7a8303 Minor cleanup of the phonon. Default value of flmixdpot set in a single place.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5410 c92efa57-630b-4861-b058-cf58834340f0
2009-02-14 06:33:04 +00:00
giannozz 017abc0fd8 Harmonization of the three h*psi routines (two were contain-ed, one
was not: now none is contain-ed)


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5409 c92efa57-630b-4861-b058-cf58834340f0
2009-02-13 16:11:32 +00:00
giannozz 77396eb935 Misspells
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5408 c92efa57-630b-4861-b058-cf58834340f0
2009-02-13 12:12:36 +00:00
giannozz 21a34bf9d0 tqr no longer experimental - it works
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5407 c92efa57-630b-4861-b058-cf58834340f0
2009-02-13 12:11:53 +00:00
giannozz 89d2fb1433 Compilation problem after last addition
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5406 c92efa57-630b-4861-b058-cf58834340f0
2009-02-13 12:11:14 +00:00
paulatto 5dd3bdca67 Realus fixed again (second derivative in first point is not first derivative
in second point), actually it makes almost no difference.
LP


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5405 c92efa57-630b-4861-b058-cf58834340f0
2009-02-12 18:15:07 +00:00
paulatto 761dd704e2 For real-space augmentation only:
For pseudopotentials not using analitical expression of aug.charge close to
nuclei (i.e. all those generated with ld1.x), random meaningless numbers
were feed to the spline interpolators for the points closer to the nucleus.

This caused unexpected "wrong charge" errors and low accuracy. Agreement
with G-space algorithm is now increased 10 times.

LP


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5404 c92efa57-630b-4861-b058-cf58834340f0
2009-02-12 17:32:34 +00:00
paulatto c6b7fadf58 Added PAW (and UPF with l-dependent augmentation) support for real-space
augmentation. Fixed an harmless but annoying bug that caused the coed to
access upf%qfcoef even when they were not used.
LP


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5403 c92efa57-630b-4861-b058-cf58834340f0
2009-02-12 14:53:45 +00:00
dalcorso 5f6123d88b Futher comments and cleanup.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5402 c92efa57-630b-4861-b058-cf58834340f0
2009-02-12 09:08:46 +00:00
dalcorso c81fb61ef2 Some comments in the phonon code updated. References to the formulas in notes
substituted with references to the formulas in PRB 64, 235118 (2001).


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5401 c92efa57-630b-4861-b058-cf58834340f0
2009-02-12 08:56:50 +00:00
giannozz 0be50d26ca Dependencies updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5400 c92efa57-630b-4861-b058-cf58834340f0
2009-02-12 08:24:03 +00:00
giannozz 5292d83a82 Unused routines removed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5399 c92efa57-630b-4861-b058-cf58834340f0
2009-02-12 08:22:16 +00:00
degironc d3ebcd964d additions to makefile files needed by the new DFT+DMFT part
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5398 c92efa57-630b-4861-b058-cf58834340f0
2009-02-12 08:09:57 +00:00
degironc 83f2bfae65 new files and modifications needed to extract a TB hamiltonian in order to
perform a DFT+DMFT calculation.
contributed by Dmtry Korotin, V.I. Anisimov, N. Binggeli et al.
The DMFT part will be loaded separately into QE-forge.

SdG


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5397 c92efa57-630b-4861-b058-cf58834340f0
2009-02-12 08:07:11 +00:00
giannozz 9131c8887c Explicit statement that unshifted grids are needed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5396 c92efa57-630b-4861-b058-cf58834340f0
2009-02-10 17:36:15 +00:00
giannozz 99db3df946 Minor change and suggestions by Karoly Nemeth about pw2casino
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5395 c92efa57-630b-4861-b058-cf58834340f0
2009-02-10 15:18:50 +00:00
giannozz 58b73a974f Real-space calculation of Q(r) had a bug that becomes apparent only with v.2
UPF pseudopotentials (qfunc not allocated for non-ultrasoft pp)


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5394 c92efa57-630b-4861-b058-cf58834340f0
2009-02-10 15:15:30 +00:00
seitsonen 2130704503 Moved the initialisation of 'paw_recon[nsp]' into a new location where it
is more logical and easier to catch errors. Thanks to error report from
Christos Gougoussis and Matteo Calandra (IMPMC, Paris).

       apsi = Ari P Seitsonen


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5393 c92efa57-630b-4861-b058-cf58834340f0
2009-02-09 17:01:29 +00:00
giannozz 758e688875 Less alarmistic message when atomic wavefunction are renormalized.
Reminder: never ever write "WARNING" on output without a good reason


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5392 c92efa57-630b-4861-b058-cf58834340f0
2009-02-09 16:58:31 +00:00
dalcorso f0212fe24c Problem with compilation.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5391 c92efa57-630b-4861-b058-cf58834340f0
2009-02-09 08:27:54 +00:00
dalcorso 9de3f198a0 Bug fix: missing broadcast of the frequencies.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5390 c92efa57-630b-4861-b058-cf58834340f0
2009-02-08 17:48:46 +00:00
dalcorso 07d2f0d835 Phonon cleanup. The reading of the recover file is done by a single routine.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5389 c92efa57-630b-4861-b058-cf58834340f0
2009-02-08 17:27:46 +00:00
dalcorso ec97e29ebd Bug fix: the phonon recover with modenum /= 0 was not working any more.
This bug was introduced in the cvs version on 01 Sep, 2008.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5388 c92efa57-630b-4861-b058-cf58834340f0
2009-02-08 17:14:59 +00:00
dalcorso 07f7a868e7 Bug fix: the phonon recover with US-PPs was not working any more. This bug was
introduced in the cvs version on Sep. 18, 2008.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5387 c92efa57-630b-4861-b058-cf58834340f0
2009-02-08 17:06:29 +00:00
giannozz 8cab67540f First attempt towards an easier to plot format for atomic pseudo-wavefunctions.
Not sure it works in all cases, though. The format of file_wavefunctionsps
is basically unchanged, but all-electron wavefunctions are added in the
right side of the file, with the same ordering and lables


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5386 c92efa57-630b-4861-b058-cf58834340f0
2009-02-06 18:17:17 +00:00
giannozz 3649fd5ac8 Last commit broke metadynamics - should be fixed now. Anybody knowing
what the option 'phase_space' is supposed to do, and whether it can be
used independently of metadynamics?


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5385 c92efa57-630b-4861-b058-cf58834340f0
2009-02-05 16:17:45 +00:00
giannozz 5367f42e42 More explicit error message
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5384 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 20:57:48 +00:00
giannozz e3cc136770 Spelling errors
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5383 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 20:54:44 +00:00
giannozz 6817a11771 Updated - please do not forget to mention fixed bugs here!
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5382 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 18:16:41 +00:00
dalcorso 9ee2377238 Cleanup of the phonon code. In the noncollinear case, the spin
indeces are calculated only once, not in every routine.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5380 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 17:18:31 +00:00
giannozz b6a6652874 starting_magnetization should not exceed +/- 1
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5378 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 15:10:21 +00:00
paulatto bd70186e0b Detailed specification of previous bug-fix.
LP


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5376 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 14:14:10 +00:00
dalcorso 44ea523f33 Bug fix: problem with effective charges d Force / d E in the
noncollinear+NLCC case.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5375 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 13:44:47 +00:00
paulatto 5de35fd8a5 In the tefield case, code assumed external electric field parallel to z axis
all the other cases produced grossly wrong results.
LP


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5374 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 13:44:45 +00:00
giannozz 0695a6cc02 Documentation updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5372 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 11:08:39 +00:00
dalcorso c2bdfb0086 Further cleanup of the phonon code. The indeces of k and k+q are no more
recalculated in every routine.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5371 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 10:25:03 +00:00
giannozz 27c6cdf10e 1) Non scf calculations use atomic positions from the data filem no longer
those provided in input. I think that this is the correct behaviour and that
the previous one was a source of trouble and of useless additional work.
I cannot see any reason why one should perform a non scf calculation with
one set of atomic positions and the scf potential relative to a different one.
2) After a reshuffling of input.f90 (which has become a mess beyond control),
the check on starting_magnetization works again. input.f90 should be
rewritten like the corresponding routine of CP. 3) Dependencies updated


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5370 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 09:41:18 +00:00
dalcorso f1d1646946 Small cleanup.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5369 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 08:36:52 +00:00
dalcorso 1d2ceae531 Minor cleanup of the phonon code (part 2). USE phcom and USE pwcom
substituted by multiple USE statements.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5368 c92efa57-630b-4861-b058-cf58834340f0
2009-02-04 08:28:27 +00:00
dalcorso 8deedeccf7 Minor cleanup of the phonon code. USE phcom substituted with multiple USE statements
in many routines.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5367 c92efa57-630b-4861-b058-cf58834340f0
2009-02-02 10:52:58 +00:00
dalcorso 981953dde5 Minor cleanup.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5366 c92efa57-630b-4861-b058-cf58834340f0
2009-02-02 08:35:23 +00:00
dalcorso 14f4b86213 Bug fix: missing deallocation.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5365 c92efa57-630b-4861-b058-cf58834340f0
2009-02-02 08:25:21 +00:00
giannozz 4380fca2e7 Unused routines removed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5364 c92efa57-630b-4861-b058-cf58834340f0
2009-02-01 15:50:42 +00:00
giannozz 3b55e4d268 Uninitialized variables related to 1/r pseudopotentials
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5363 c92efa57-630b-4861-b058-cf58834340f0
2009-02-01 14:27:53 +00:00
cavazzon 221ce10bf1 - bug fix: there was a synchronization problem between processor groups
associated with different images (ex. NEB) when the code was compiled for
  SCALAPACK.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5362 c92efa57-630b-4861-b058-cf58834340f0
2009-01-31 14:42:40 +00:00
dalcorso 41417085e6 Problem with the Fermi energy shift in the noncollinear magnetic case.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5361 c92efa57-630b-4861-b058-cf58834340f0
2009-01-30 14:52:09 +00:00
dalcorso 7917ca98c3 Missing deallocation in solve_e in the noncollinear/spin-orbit case.
Cleanup.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@5360 c92efa57-630b-4861-b058-cf58834340f0
2009-01-30 14:47:35 +00:00