Commit Graph

4288 Commits

Author SHA1 Message Date
giannozz 1b07c49f99 Minor changes
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4399 c92efa57-630b-4861-b058-cf58834340f0
2007-11-06 10:49:07 +00:00
degironc 79f0fd08c0 more scf_mod changes
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4398 c92efa57-630b-4861-b058-cf58834340f0
2007-11-06 10:26:07 +00:00
paulatto 8c462e6c22 PAW energy can get very high for heavy atoms, it's better to use two more spaces to print it.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4397 c92efa57-630b-4861-b058-cf58834340f0
2007-11-05 18:05:47 +00:00
dalcorso 0cd0c1b939 Makefile update and small correction to previous commit.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4396 c92efa57-630b-4861-b058-cf58834340f0
2007-11-05 17:38:31 +00:00
dalcorso 87f686feb8 Introduced a check to stop all post-processing codes when okpaw is true.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4395 c92efa57-630b-4861-b058-cf58834340f0
2007-11-05 17:11:28 +00:00
dalcorso d3a643f591 Bug fix: a wrong allocation.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4394 c92efa57-630b-4861-b058-cf58834340f0
2007-11-05 17:01:47 +00:00
fratesi 891215524d Added a tool for merging "prefix.path" files. (GF)
USAGE:  path_merge.bash  file0.path n0 m0  file1.path n1 m1

    merges the two path files file0.path and file1.path by taking the
    images n0 to m0 from file0.path and images n1 to m1 from file file1.path.
    Merged file is written to standard output.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4393 c92efa57-630b-4861-b058-cf58834340f0
2007-11-05 16:58:52 +00:00
giannozz c5f14b460d Calculation of atomic wavefunctions merged into a single routine
Preliminary operations aimed at introducing yet another choice of
initial wavefunctions


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4392 c92efa57-630b-4861-b058-cf58834340f0
2007-11-05 13:39:28 +00:00
dalcorso 7fce65f9ac Bug fix: when the number of processors was larger than the number of atoms
the program was not working in parallel in the paw case.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4391 c92efa57-630b-4861-b058-cf58834340f0
2007-11-05 10:12:00 +00:00
degironc c684e80af1 scf_mod update
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4390 c92efa57-630b-4861-b058-cf58834340f0
2007-11-04 20:28:20 +00:00
degironc 349713ce63 Harris-Weinert-Foulkes energy was occasionally incorrectly calculated at the first interation.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4389 c92efa57-630b-4861-b058-cf58834340f0
2007-11-04 07:58:43 +00:00
giannozz 8370a4d99b more smd removal
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4388 c92efa57-630b-4861-b058-cf58834340f0
2007-11-02 14:38:04 +00:00
giannozz 66cae3b1cc More smd removal
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4387 c92efa57-630b-4861-b058-cf58834340f0
2007-11-02 14:30:06 +00:00
degironc 4281a2b252 more changes in scf_mod
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4386 c92efa57-630b-4861-b058-cf58834340f0
2007-11-02 13:39:38 +00:00
giannozz bd21d1a21d More tests
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4385 c92efa57-630b-4861-b058-cf58834340f0
2007-11-01 15:39:51 +00:00
giannozz a09a87f88d tests updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4384 c92efa57-630b-4861-b058-cf58834340f0
2007-11-01 15:03:11 +00:00
giannozz c1d8cf40fc Strings Method ('smd') removed from CP and from examples.
I am not sure whether smd in PW is working or not


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4383 c92efa57-630b-4861-b058-cf58834340f0
2007-11-01 13:39:43 +00:00
degironc fab8c49e9c scf_mod again..
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4382 c92efa57-630b-4861-b058-cf58834340f0
2007-11-01 11:19:31 +00:00
degironc 93c72b5f3b make.depend again and a corrected scf_mod.f90
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4381 c92efa57-630b-4861-b058-cf58834340f0
2007-10-31 23:02:23 +00:00
degironc eae651d0e9 missing module...
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4380 c92efa57-630b-4861-b058-cf58834340f0
2007-10-31 15:52:50 +00:00
degironc d5387129e5 module scf has been separated from PW/pwcom.f90 (put in PW/scf_mod.f90) in
order to reorganize the variables needed for mixing and scf cycle.
More changes will follow....


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4379 c92efa57-630b-4861-b058-cf58834340f0
2007-10-31 14:28:18 +00:00
degironc 33a9eac8b4 make.depend updates + a couple of small additions in PW/rad_paw_routines and atomic/c6_tfvw
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4378 c92efa57-630b-4861-b058-cf58834340f0
2007-10-31 10:31:55 +00:00
giannozz 70b998cd43 More Berry cleanup
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4377 c92efa57-630b-4861-b058-cf58834340f0
2007-10-30 09:19:39 +00:00
giannozz 8f4019172a The Ylm used in Berry's phase calculation were inconsistent with the
Clebsh-Gordan coefficients, calculated for a different set of Ylm.
Moreover the call to the routine calculating Ylm was incorrect.
The error was small (it affected only the US part with l>0) but not
negligible. I don't know if it is fixed now, but it wasn't correct
before for sure. Berry's phase calculation extended to L=3.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4376 c92efa57-630b-4861-b058-cf58834340f0
2007-10-30 08:18:46 +00:00
giannozz cc6f92ed91 Incorrect behavior in parallel in some cases
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4375 c92efa57-630b-4861-b058-cf58834340f0
2007-10-29 15:04:25 +00:00
giannozz 76aafc384e Added the possibility to perform calculations using the true Coulomb
potential instead of a pseudopotential. May or may not work! only for pw.x
and only for people knowing what they are doing. It requires the generation
of a fake UPF pseudo containing a radial grid, atomic wavefunctions, little
else, using the modified version of ld1.x (option "write_coulomb=.true.").


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4374 c92efa57-630b-4861-b058-cf58834340f0
2007-10-29 13:47:45 +00:00
giannozz 4e55ebdfaa Redundant variable nlcc eliminated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4373 c92efa57-630b-4861-b058-cf58834340f0
2007-10-29 10:20:01 +00:00
giannozz 02f32822e5 Automated tests updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4372 c92efa57-630b-4861-b058-cf58834340f0
2007-10-29 10:13:29 +00:00
giannozz 22f5ad8648 Clocks are now safe in parallel mode (i.e. even if some of them are not
started nothing nasty happens)


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4371 c92efa57-630b-4861-b058-cf58834340f0
2007-10-26 19:11:54 +00:00
giannozz 97177f9531 Version number updated + some cleanup
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4370 c92efa57-630b-4861-b058-cf58834340f0
2007-10-26 16:47:09 +00:00
umari 8f07f632c8 Some cleaning 8-->dp, and dynamic allocations introduced
P.U.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4369 c92efa57-630b-4861-b058-cf58834340f0
2007-10-26 00:48:51 +00:00
umari 1850e1afd8 Small bug recently introduced
as upf%qqq were not working with efield
I commented  those lines, just wainting to be re-introduced

P.U.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4368 c92efa57-630b-4861-b058-cf58834340f0
2007-10-25 20:26:14 +00:00
giannozz 7d707a0d4b Duplicated line causing compilation error removed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4367 c92efa57-630b-4861-b058-cf58834340f0
2007-10-25 08:57:56 +00:00
giannozz 84a9283679 This was for sure a bug - all redundant variables have to be initialized
(until they are removed)


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4366 c92efa57-630b-4861-b058-cf58834340f0
2007-10-24 16:02:51 +00:00
giannozz a2f2848f92 More static arrays and dimensions removed: rho_at rho_atc natx nbndxx
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4365 c92efa57-630b-4861-b058-cf58834340f0
2007-10-24 15:36:52 +00:00
ceresoli 45338d9069 Small changes, in order to compile GIPAW. (D.C.)
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4364 c92efa57-630b-4861-b058-cf58834340f0
2007-10-24 09:16:57 +00:00
giannozz 9683c28102 make.depend files updated
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4363 c92efa57-630b-4861-b058-cf58834340f0
2007-10-23 20:06:04 +00:00
giannozz 8df6e5b3a5 More removal of static arrays: chi, nchi, lchi, jchi, oc, nchix
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4362 c92efa57-630b-4861-b058-cf58834340f0
2007-10-23 19:47:26 +00:00
degironc 865dc88598 updated documentation for constrained_magnetization
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4361 c92efa57-630b-4861-b058-cf58834340f0
2007-10-23 14:55:49 +00:00
degironc aae083bf9e constrained_magnetizatio='atomic' has been extended to the collinear case (nspin=2), where only the z-component of the local moment is constrained.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4360 c92efa57-630b-4861-b058-cf58834340f0
2007-10-23 14:51:28 +00:00
dalcorso cb65aa6246 A missing DP was producing random NaN in the paw case.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4359 c92efa57-630b-4861-b058-cf58834340f0
2007-10-23 13:38:05 +00:00
paulatto d858b4a51b New PAW version, now completely working on radial grid. Code on FFT grid can still be compiled using __GRID_PAW flag (but may not work a
nymore). Other things working: parallel, gamma-only.

Files vxc_t, exc_t and vxcgc moved from atomic to Modules (to prevent cyclic dependencies).

Other random fixes: a kind in init_vloc, a call to infomsg in ./PH/add_for_charges.f90.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4358 c92efa57-630b-4861-b058-cf58834340f0
2007-10-22 14:54:29 +00:00
dalcorso 163557457a Small changes.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4357 c92efa57-630b-4861-b058-cf58834340f0
2007-10-21 15:20:24 +00:00
dalcorso 4bb6d656a5 The routine ascheqps substituted with a more precise routine.
For a while the old routine is still there for tests. Just call
ascheqps_old instead of ascheqps in ascheqps_drv.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4356 c92efa57-630b-4861-b058-cf58834340f0
2007-10-21 14:59:56 +00:00
dalcorso 52782c136f Cleanup.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4355 c92efa57-630b-4861-b058-cf58834340f0
2007-10-21 14:52:31 +00:00
dalcorso 2dce733c55 Problems with paw after recent changes to the pseudopotential variables.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4354 c92efa57-630b-4861-b058-cf58834340f0
2007-10-20 12:14:35 +00:00
giannozz 26628f09a8 A few changes lost in last commit re-installed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4353 c92efa57-630b-4861-b058-cf58834340f0
2007-10-18 15:12:38 +00:00
seitsonen 0f8940bd08 Calculation of the chemical shift due to the core electrons added; only
tested on carbon atoms so far

       apsi


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4352 c92efa57-630b-4861-b058-cf58834340f0
2007-10-18 13:21:54 +00:00
umari 77aec8c1f1 Now Berry's phase electric fields work also in parallel
for gdir == 1 and gdir ==2

P.U.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4351 c92efa57-630b-4861-b058-cf58834340f0
2007-10-17 19:12:51 +00:00
giannozz f7cc9bfa32 Fixed incorrect allocation of auxiliary array introduced in last commit
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4350 c92efa57-630b-4861-b058-cf58834340f0
2007-10-17 10:24:59 +00:00