scnc
/
quantum-espresso
mirror of https://gitlab.com/QEF/q-e.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
10,974 Commits 5 Branches 40 Tags 1.8 GiB
Commit Graph

31 Commits

Author SHA1 Message Date
giannozz 25a3d719e7 cgather_sym replaced by generic interface gather_grid - there is no reason 
to collect a FFT array on all processors as cgather_sym does, since only the 
first processor writes to file


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@11532 c92efa57-630b-4861-b058-cf58834340f0
 2015-05-08 12:55:32 +00:00
giannozz 7c01062ba0 Final (?) set of changes to the build system, especially for Windows (Axel) 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10590 c92efa57-630b-4861-b058-cf58834340f0
 2013-11-07 21:06:06 +00:00
giannozz c1f947ba6e make.depend updated + minor format fix 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10568 c92efa57-630b-4861-b058-cf58834340f0
 2013-11-03 20:19:48 +00:00
giannozz 3fcc01b467 Variables for task group (actually a single one) moved to where it belongs 
(I think): mp_bands.f90 . Many changes but nothing dangerous. Note that 
codes not in svn may be broken by this change, but the fix is very simple


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10567 c92efa57-630b-4861-b058-cf58834340f0
 2013-11-03 19:16:37 +00:00
giannozz 95600c8582 More cleanup of parallel modules: mp_global no longer contains references to 
variables for the "world" MPI communicator. The latter are to be found in 
world_comm instead. mp_global should be used only to start and to end the
various parallelization levels. Many small but harmless changes: a few 
variables removed or moved to another module in most cases.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10558 c92efa57-630b-4861-b058-cf58834340f0
 2013-10-27 09:09:10 +00:00
ccavazzoni b1fabb0fd9 - more clean-up about mp call without communicator 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10498 c92efa57-630b-4861-b058-cf58834340f0
 2013-09-28 23:32:46 +00:00
ccavazzoni 8b9c1a54d0 - more Axel patches for LAMMPS and QE integration 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10493 c92efa57-630b-4861-b058-cf58834340f0
 2013-09-27 14:37:51 +00:00
ccavazzoni 3e09ccb6ba adding explicit communicator in "mp" cals 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10485 c92efa57-630b-4861-b058-cf58834340f0
 2013-09-26 20:49:26 +00:00
giannozz e0086b6882 The IFLAGS variable is set in make.sys to a value that works if sources are 
one directory down, not two directories down. Now it is explicitly set to
something (hopefully) appropriate everywhere. Note that include/f_defs.h 
is obsolete and shouldn't be included (it is still in GWW and TDDFPT).


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10446 c92efa57-630b-4861-b058-cf58834340f0
 2013-08-26 16:36:27 +00:00
dalcorso 1863022f1e D3 updated to the new convention on the record lenghts. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10168 c92efa57-630b-4861-b058-cf58834340f0
 2013-04-19 13:48:26 +00:00
paulatto 2a1be24082 In teh case where then umber of crystal symmetry is > 1, but the small group of q has no symmetry, the code was regenerating the patterns instead of reading them from file. 
This is wrong and could cause Input/Output errors or even wrong result (in some very rare case, however)



git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@10127 c92efa57-630b-4861-b058-cf58834340f0
 2013-04-09 11:51:48 +00:00
giannozz 883ec9c5d3 Misc documentation and make.depend updates 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@9667 c92efa57-630b-4861-b058-cf58834340f0
 2012-11-23 20:06:09 +00:00
dalcorso 6e9e3a6fff Inserted a check to stop the d3 program if pw.x and ph.x were run with the 
nband flag. Task groups disabled.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@9665 c92efa57-630b-4861-b058-cf58834340f0
 2012-11-23 16:41:37 +00:00
paulatto 1ae53585d6 Problem with dfile_name only doing i/o in ionode solved. 
io_pattern is now only called by ionode, when reading the variables it reads are manually broadcast.

The removal of io_pattern from set_d3irr should be harmless as modenum is always set to zero in the d3 code.



git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@9355 c92efa57-630b-4861-b058-cf58834340f0
 2012-08-21 14:30:44 +00:00
dalcorso a6d9dcea0f make_depend updated. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@9247 c92efa57-630b-4861-b058-cf58834340f0
 2012-07-29 19:10:11 +00:00
giannozz 17edd139e6 Updated make.depend 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@9135 c92efa57-630b-4861-b058-cf58834340f0
 2012-06-28 18:15:43 +00:00
paulatto 5ecf818d76 d3 code was not compiling any more after amu_ry 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@9122 c92efa57-630b-4861-b058-cf58834340f0
 2012-06-22 10:16:19 +00:00
giannozz 801db10bfb Cleanup of replicated constants: amconv => amu_ry 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@9117 c92efa57-630b-4861-b058-cf58834340f0
 2012-06-20 09:17:35 +00:00
paulatto 5e525b1ea3 Fixed possible source of troubles where d3 would add non-lineat core correction to fildrho and fild0rho file but never remove it. This caused the calculation to produce wrong results if the same fildrho file was used more then once. Now the core correction is added to a separate file. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8722 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-24 11:29:05 +00:00
giannozz 38e798eed7 h_psiq hack: I think it is preferrable to give to igkq the correct value 
in the external loop over k-points, rather than passing it as an argument.
Just two lines in TD-DFPT needed. Routine commutator_Hx_psi, used only by
linear-response code, moved to PHonon: no reason to have it in PW


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8659 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-15 20:31:07 +00:00
paulatto 7450e89b83 d3 fixed 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8651 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-15 10:25:11 +00:00
giannozz 7bd937ab3d More make.depend forgotten in previous commit updated 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8636 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-13 20:31:15 +00:00
paulatto 9474207b5c io_pattern now takes an extra argument: the xq point. It is written and/or read from the pattern file. This is necessary for the d3 code but also allows a few sanity checks in other codes (the relative changes have slipped in the previous commits). 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8628 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-13 15:22:34 +00:00
paulatto 02d807788f seqopn moved to Modules/io_files.f90 not only to keep company to diropn but also to have an explicit interface. This is necessary to introduce an optional argument to specify the directory (as it had been done for davcio some time ago). This is also necessary for the d3 code, but can also help a finer grained distribution of temporary files. 
LP




git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8624 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-13 15:12:29 +00:00
paulatto 5a7c2bf3ff set_drhoc modified to accept the array drc as input instead of taking it from the module, as it already does for xq. 
This allows for some semplification of the d3 code, and possibly other codes

LP



git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8623 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-13 15:09:31 +00:00
paulatto eebb8de9ec I've modified h_psiq in order to accept one more argument (the igkq mapping) and get rid of those if/then conditions. 
I would also suggest we should try really hard to NOT introduce this kind of hacks, especially not in core subroutines.



git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8622 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-13 15:06:05 +00:00
obm 0bc6eeb0bd open64 compatibity: #warning compiler directive is depreciated (in general), and it is not supported by open64 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8585 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-07 15:55:35 +00:00
giannozz 4de42518ea No reason to have both __PARA and __MPI preprocessing options: they have the same meaning. __MPI is more explicit, so I am keeping it 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8572 c92efa57-630b-4861-b058-cf58834340f0
 2012-02-02 22:12:26 +00:00
marsamos 59d6818a9b PW/src PW/tools compiling makedeps updated 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8335 c92efa57-630b-4861-b058-cf58834340f0
 2011-12-14 17:33:33 +00:00
marsamos 92c87720b0 path error in d3.x and phcg.x links . 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8169 c92efa57-630b-4861-b058-cf58834340f0
 2011-09-27 11:54:49 +00:00
marsamos 2496b745bd PH re-structured. PH Now called PHonon, D3 and Gamma moved to dir PHonon 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@8158 c92efa57-630b-4861-b058-cf58834340f0
 2011-09-21 13:43:24 +00:00