scnc
/
quantum-espresso
mirror of https://gitlab.com/QEF/q-e.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
15,665 Commits 5 Branches 40 Tags 1.8 GiB
Commit Graph

261 Commits

Author SHA1 Message Date
Samuel Ponce 51e3f0d079 Corrections regarding elec self-energy restart option. 2018-10-17 10:33:17 +01:00
Samuel Ponce 130204d63e Bug in parallel raised by test-farm 2018-10-16 22:01:59 +01:00
Samuel Ponce f5ff40b4e1 Debug for serial case. 
IBTE must be compiled with MPI.
 2018-10-16 18:12:52 +01:00
Samuel Ponce dd817a575a Cleaning and optimization in EPW 
1) Complete re-writing of the IBTE (at least an order of magnitude faster).
2) Supporting both k-point symmetry and no-symmetry
3) Creating of a few new modules
4) Addition of a epw_mob_ibte_sym for testing the IBTE using k-point symmetry.
Both epw_mob_ibte_sym and epw_mob_ibte should give the same result but
epw_mob_ibte_sym is much faster.
 2018-10-16 17:06:08 +01:00
Samuel Ponce 1cf7322507 Update epw test-farm 
(see previous commit).
 2018-10-11 16:05:38 +01:00
Samuel Ponce 50c0cc93e8 Update the way WS is constructed. 
If use_ws == .true. use wannier_centers and atomic positions
If use_ws == .false. use zone centered Wigner-Seitz. This should
recover the behavior of EPW from QE 6.2.1 and before.
Thefore I have to update benchmark file (other benchmark directory will follow).
 2018-10-11 14:54:17 +01:00
Samuel Ponce 25ed0d07a2 Debug raised by test-farm 
Imposition of the phase (input variable lphase) is crucial to obtain
consistent results between etf_mem 1 and 2.
lphase has therefore been added to the input of the epw_polar test.
 2018-09-22 18:00:43 +01:00
Samuel Ponce e81850cabe New implementation of the WS vectors in EPW 
The new implementation relies on Wannier centers positions and is
triggered with the input variable "use_ws == .true."
When false, recover the old behavior.
No test has been added yet for this.

The update of the epw_polar test epw2 and epw3 is because of the
removal in EPW of the exponential look-up table which slightly change
numerical results.
 2018-09-21 11:42:24 +01:00
Paolo Giannozzi 1d34637117 Upon suggestion by Samuel: relative tolerance on phonon frequencies increased 
for EPW test. This should get rid of annoying messages from buildbot.
 2018-09-11 19:34:41 +02:00
Samuel Ponce 5886c5b30f Increase of IBTE testing parameters. 2018-08-30 10:08:08 +01:00
giannozz dc2c1db2a9 Merge branch 'develop' into 'develop' 
Real-space Q(r) plus misc. fixes

See merge request QEF/q-e!166
 2018-08-29 21:01:18 +00:00
Samuel Ponce 72a8f3c7d6 Correct some circular dep. in EPW 2018-08-29 17:28:35 +01:00
Samuel Ponce cec8d38b08 Updated tests following the recent change in sym. 
Symmetries are now detected in the case of frac. translation
and the FFT grid is commensurate.
 2018-08-29 15:28:25 +01:00
Paolo Giannozzi c2c1d895bc yet another small increase of the threshold to avoid annoying bogus 
messages from buildbot
 2018-08-29 12:08:01 +02:00
Paolo Giannozzi caf21dd387 Treatment of fractional translations significantly changed. Before: 
- operations with fractional translations were discarded if not commensurate
  with FFT grid (unless input variable "use_all_frac" was set to true)

Now:
- FFT grids are made commensurate with fractional translations
  (unless use_all_frac is set to true)

Advantages:
- the number of k-points and of symmetry operations no longer depends upon the
  cutoff via the FFT grid (could happen, and did happen, before)
- cleaner code, allowing easier automatic parallelization over k-points
Disadvantage:
- the FFT grid may become slightly larger

Changes are minimal because the machinery was already in place (see variable
"fft_fact" that contains factors for FFT dimensions). Cleanup: useless calls
to fft_type_allocate removed (it is called by fft_init_type anyway).
Documentation and examples updated. The usage of "use_all_frac" is a little
bit weird, though (should be the other way round).
 2018-08-26 21:24:12 +02:00
Paolo Giannozzi 6bd57ac2f3 [these changes should have been present in a previous commit, but were not] 
More removal of old I/O: input_drho, output_drho now use new format. Not sure
they work, though (no test available) and not sure they are used (the previous
version was broken for sure). Fixed a bug added in previous commit.
 2018-08-16 09:04:35 +02:00
Paolo Giannozzi 2e6e2e27e1 More removal of old I/O: input_drho, output_drho now use new format. Not sure 
they work, though (no test available) and not sure they are used (the previous
version was broken for sure). Fixed a bug added in previous commit.
 2018-08-15 22:02:15 +02:00
Paolo Giannozzi 0381d14dd1 Kinetic energy density is now written in G-space as well. It seems to work 
but no warranty. Tests for band structure with metaGGA added.
 2018-08-15 09:54:05 +02:00
Paolo Giannozzi 45fc53b530 Merge branch 'develop' of gitlab.com:QEF/q-e into develop 2018-07-18 18:04:37 +02:00
Paolo Giannozzi 93ea91abbc Minor updates 2018-07-18 17:47:59 +02:00
Paolo Giannozzi 232a2fbe57 Reference to SVN in the names of test-suite benchmark files and in the 
developer manual removed. Maybe it would be preferrable to have benchmarks
from a given reference version?
 2018-07-18 15:15:46 +02:00
Miguel Carvajal 68ee7b5e87 Rename module wavefunctions_module to wavefunctions 2018-07-12 16:11:48 +02:00
Samuel Poncé 2cc17ae28f Add the possibility of linear mixing for IBTE. 2018-07-08 16:06:15 +01:00
Samuel Ponce 66eed22682 Change -\tau to +\tau for wigner_g 
Update EPW test-farm.
 2018-07-03 15:19:16 +01:00
Samuel Ponce bc2564cb8f Small bug in size of passed array. 2018-07-02 18:23:42 +01:00
Samuel Poncé ddd1548796 One file was not removed in sequential 
Also increase logavg tol
 2018-07-02 09:49:36 +01:00
Samuel Ponce 617e253f43 Small tweaking of tolerence for the test-farm 2018-07-01 21:14:49 +01:00
Samuel Poncé 9c5d745d2c Update on the phonon averaging on modes. 2018-07-01 18:54:26 +01:00
Samuel Poncé e8644a483b Average on lambda and gamma when phonon freq are degenerate. 2018-07-01 17:26:20 +01:00
Samuel Poncé 8cb97b078b Missing one test. 2018-07-01 16:32:57 +01:00
Samuel Poncé 1e16c0491f Update of the EPW test-suite following the new WS vectors 
Also correct some small bugs.
 2018-07-01 16:28:40 +01:00
Paolo Giannozzi 36d90d239a Gamma test for b3lyp now tests stress as well 2018-06-23 18:31:40 +02:00
Paolo Giannozzi 886b33edfb Test for spin-polarized B3LYP (O atom) replaced with a more stable one (OH) 2018-06-22 16:11:09 +02:00
Samuel Ponce f71ebb7b3d Added comment. 2018-06-21 11:33:17 +01:00
Samuel Ponce 7dc77cf02d Addition to the CINECA VM Builbot files. 
Might be useful if someone wants to create its own
test-farm or if the machine dies and we loose the setup files.
 2018-06-21 11:19:36 +01:00
Samuel Ponce a36eb13853 Release-notes updated and tweak of test-suite. 2018-06-21 11:04:52 +01:00
Samuel Poncé 70e6f56820 Erroneous call to transport_coeff when doing superconductivity. 
It should also now be possible to do IBTE with a fixed Fermi level
(typically in the middle of the bandgap)
and two floating Fermi level for electron and hole mobility.
 2018-06-20 16:30:24 +01:00
Samuel Poncé ee4fc82d24 Debug dme in parallel 
Update IBTE benchmark.
Extent velocity symmetrization.
 2018-06-20 13:34:14 +01:00
sponce 92e0c57a84 Velocity symmetrisation in IBTE. 2018-06-19 18:09:22 +01:00
sponce 800f1da781 Addition of timing information to mobility 
And teaking of mobility test-farm.
 2018-06-19 15:54:53 +01:00
sponce 616ace204c Correct from ecut 15 to 20 in epw_mob/nscf 
Sligthly increase tol for mobility to speedup
calculations on the test-farm.
 2018-06-19 09:41:32 +01:00
sponce 8aac573141 Remove the /tmp/save environment variable as it 
might clash when different bots are running.
Suggested by P. Gioannozzi
 2018-06-18 17:22:01 +01:00
Samuel Ponce a0281acc6c Tweaking the test-suite to pass on Buildbot for EPW 2018-06-18 17:00:59 +01:00
Samuel Ponce b797f4bb4e Debug a summing problem in displayed mobility 
in parallel.
 2018-06-18 14:17:13 +01:00
Samuel Ponce 1f237eaacf EPW test-suite 
1) Addition of testing for IBTE
2) Speedup the epw_polar test by reduction of ecut from 60 Ry to 20 Ry
3) Update of the epw_mob tests because of change in printing format
 2018-06-18 10:59:37 +01:00
Samuel Ponce fcec46fe25 Debug of the epw_base test-suite test. 2018-06-17 17:14:37 +01:00
Paolo Giannozzi 310a91b86f Test of ibrav and lattices modified so that it can be run with 
linear-algebra parallelization on 4 processors
 2018-06-15 14:12:25 +02:00
Paolo Giannozzi b65bb22cb3 Error in documentation file for PW fixed 
There was a reason why the various "include" in test-suite had never previously
been replaced by the correct "source" command: they were not working. My
apparently innocuous change completely broke test-suite.
 2018-06-05 10:45:16 +02:00
Paolo Giannozzi 053111dcea "include" replaced by "source" in Makefiles 2018-06-01 14:17:46 +02:00
Paolo Giannozzi bcb4ff3cdb Oops ... mispell 2018-05-10 19:10:27 +02:00
Paolo Giannozzi 58705a9216 xdm test merged with other vdw-related tests 2018-05-09 18:31:07 +02:00
Paolo Giannozzi f9779b78d4 Oops - removing conflicts 
Merge branch 'develop' of gitlab.com:QEF/q-e into develop
 2018-05-09 18:20:57 +02:00
Paolo Giannozzi aab5d38fdb DFT-D3 wasn't working with NEB due to missing deallocation. Cleanup: api.f90 
should be kept as intact as possible, QE-specific changes in dftd3_qe.f90.
Tests updated, test for dft-d3 added
 2018-05-09 12:40:22 +02:00
sponce e5285c9e47 Rename folder 
Test that should not be tested by buildbot should have the
prefix not_XX
 2018-05-08 22:12:47 +01:00
Samuel Ponce a9916f8da2 Removal of parallel_k and parallel_q flags. 
parallel_q was not used and lead to large code duplications.
 2018-05-04 17:41:45 +01:00
Samuel Ponce 656dea5cbb Add description of tests for pwscf 
Courtesy of P. Giannozzi.
 2018-05-04 17:15:36 +01:00
Samuel Ponce 51966a3a28 Removal of the depreciated "tests" folder in EPW. 
Now all the EPW tests are in q-e/test-suite.
Addition of tests description in q-e/test-suite/README.
 2018-05-04 15:21:59 +01:00
Pietro Delugas b70714dcee fixed new tests which were not working 
* removed offending pseudodir and outdir variable from input.
* moved d3 example for pw_dft-d to pw_vdw
* removed d2 example which was already present
* changed the input of old d2 example, completely equivalent matches
  with the old benchmark.
 2018-04-28 12:45:23 +02:00
Pietro Delugas 30e8751243 added test for Grimme d3 semiempirical correction for VdW 2018-04-27 18:20:22 +02:00
Paolo Giannozzi c712f641a7 Minor updates to documentation, tests, make.depend. Not-so-minor correction 
to wfcinit.f90: with the new XML file, the presence of a spurios old-style
wavefunction file could break reading wavefunctions. It seems to work now
also for restart. It relies on the correct value of the variable twfcollect
written to the xml file: must be .false. for files saved during a run,
.true. for files saved at the end of a run.
 2018-04-18 18:42:15 +02:00
Paolo Giannozzi 3a67445fdd Added check for case of LSDA with final nonmagnetic structure. 
In that case, a further run is done with nonzero initial magnetization,
to prevent falling into a local minimum zith zero magnetization.
 2018-04-16 17:35:34 +02:00
Samuel Ponce d481fe3576 Tweaking for the ind. abs. EPW test. 
Also sligthly increase lambda2 for PH
from 9.0e-2 to 9.5e-2.
 2018-04-10 12:48:10 +01:00
Samuel Ponce 94bbf550e3 Make the test faster by reducing grids. 2018-04-09 17:55:41 +01:00
Samuel Ponce ead720d9f7 New feature in EPW: Indirect absorption 
Also adding a test on that in the test-suite.
 2018-04-09 13:25:35 +01:00
sponce 8105079779 Change a relative tolerance to None in EPW tests. 2018-04-08 22:03:28 +01:00
sponce eb31b3eb02 Preventing negative phonon frequency 
when computing FD occupations. Courtesy of G. Brunin.
Update of the plasmon test following the recent debug.
 2018-04-08 11:31:26 +01:00
Samuel Ponce ab8244e14a Debug when etf_mem==2 
Update the EPW test-suite for mobility and new pp.py script.
 2018-03-27 18:16:18 +01:00
Paolo Giannozzi 60148f3e61 Oops ... 2018-03-26 07:09:43 +02:00
Paolo Giannozzi b8f5141f3d test-code creates temp directory if not existent (PW creates it, CP doesn't) 2018-03-25 20:44:28 +02:00
Samuel Ponce 714c07d37a Addition of a test for mobility. 2018-02-15 19:00:55 +00:00
Paolo Giannozzi e787597d05 Threshold too low in recently added => unstable results 2018-02-01 22:40:16 +01:00
Paolo Giannozzi 0b8a8602c4 Merge branch 'master' of https://gitlab.com/QEF/q-e 2018-01-25 09:47:03 +01:00
Stefano de Gironcoli 865173b693 run_tests_pw creates output directory if it is absent, run_tests_cp does not and crashes. 
execute run_tests_pw before run_tests_cp by default.
 2018-01-24 00:13:57 +01:00
Paolo Giannozzi 3fe16db938 Spin-polarized X3LYP, with tests - may or may not work 2018-01-23 14:18:06 +01:00
Samuel Ponce 89a13b248b Raised the lambda tolerence for EPW-pl 
From 5E-3 to 1E-2 absolute. This is because of
the recent change since one do not divide by epsilon_HEG anymore.
The the results are 12 times bigger, hence larger absolute error.
 2017-12-16 13:00:33 +00:00
Samuel Ponce cdc319bd5b Update the el-pl reference test. 2017-12-15 13:35:26 +00:00
giannozz 824d871176 Stress for meta-GGA added - Hsin-Yu Ho and Marcos Calegari Andrade 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@14013 c92efa57-630b-4861-b058-cf58834340f0
 2017-11-26 19:08:33 +00:00
sponce 48c713e765 vdw4 has been removed. 
Not supported until further notice. 


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@14003 c92efa57-630b-4861-b058-cf58834340f0
 2017-11-20 17:08:25 +00:00
sponce 6345f123f7 Correction to cope with new FFT scheme 
Courtesy of Paolo


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13980 c92efa57-630b-4861-b058-cf58834340f0
 2017-11-02 17:41:20 +00:00
sponce 7795138425 Removal of "clean" when doing run-tests-epw-parallel 
Adapting tol because of epw_soc on PGI 17 for EPW


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13929 c92efa57-630b-4861-b058-cf58834340f0
 2017-10-17 15:38:51 +00:00
sponce 7977e1e122 Adapt the tolerence for pgi 17 for EPW 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13928 c92efa57-630b-4861-b058-cf58834340f0
 2017-10-17 10:03:28 +00:00
giannozz 0253178db1 Missing space could have lead to strange problems when PARA_POSTFIX is set 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13925 c92efa57-630b-4861-b058-cf58834340f0
 2017-10-16 19:30:05 +00:00
giannozz dea199b670 Slight change to vdw4.in test (remove randomization of initial wavefunctions) 
in order to make its results, notably the final volume, more stable


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13896 c92efa57-630b-4861-b058-cf58834340f0
 2017-10-05 07:40:42 +00:00
sponce 9b218b4737 Fix extract script for pw for relaxation 
relaxation may terminate in a different amount of steps - hence we only compare
the first few (3) steps to check for consistency
then we compare the final result (structure/cell)
This solves the issues with PGI17 from the test-farm. 
Courtesy of M. Schlipf



git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13891 c92efa57-630b-4861-b058-cf58834340f0
 2017-10-03 10:47:20 +00:00
giannozz 7867722d41 testcode doesn't work with python3 (Pietro Bonfa') 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13840 c92efa57-630b-4861-b058-cf58834340f0
 2017-09-14 13:32:53 +00:00
giannozz df0a2b252e Patch by Martin Schlipd to prevent bogus test failure in case of 
- el-ph coefficients of degenerate phonons
- very small eigenvalues giving large relative errors on small absolute errors


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13838 c92efa57-630b-4861-b058-cf58834340f0
 2017-09-11 15:33:32 +00:00
giannozz b20abde1d6 Small change in threshold to prevent additional iteration due to numerical noise 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13837 c92efa57-630b-4861-b058-cf58834340f0
 2017-09-10 19:52:49 +00:00
giannozz c973340c99 The definition of the reduced-cutoff grid for EXX wasn't completely correct 
after the last round of changes, leading to discrepancies with old results
and between serial and parallel execution in some cases. USPP EXX test 
with k-points re-added: there are still some discrepancies but very small


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13835 c92efa57-630b-4861-b058-cf58834340f0
 2017-09-10 08:49:10 +00:00
sponce ca2a6740d9 Temporary removal of test uspp-hyb-k.in 
from the test-suite. Until a bugfix is found. 



git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13831 c92efa57-630b-4861-b058-cf58834340f0
 2017-09-05 10:02:24 +00:00
sponce dbe9dd6d72 Removal of the weights in the input of EPW 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13830 c92efa57-630b-4861-b058-cf58834340f0
 2017-09-05 09:02:20 +00:00
sponce 2d32977bb2 Clean more stuff for the ph tests and correct 
a bug while cleaning. 



git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13760 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-23 17:31:21 +00:00
giannozz c21657ec4d Minor test-suite changes: 
- "make compare" compares phonons as well
- add possibility to use pools for pw.x as well
- commented out annoying printout of execution line


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13726 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-19 12:45:56 +00:00
sponce bc0426f545 Debug lscreen with the lookup table. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13725 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-18 13:50:55 +00:00
sponce 1d7940613c Output of cumulant inside the main output for testing. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13724 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-18 10:17:52 +00:00
sponce 42387d944b Adding some screening capabilities 
Also cumulant post-processing.
Courtesy of C. Verdi and F. Caruso.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13723 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-18 09:40:12 +00:00
sponce 274e623de5 Debug in parallel for the new test. 
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13717 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-09 14:15:03 +00:00
sponce 44db9e3353 Addition of el-pl and according test. 
Courtesy of F. Caruso


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13716 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-09 13:26:12 +00:00
degironc 8f20e8c762 add a PARA_POSTFIX to run-cp.sh so that different parallelization 
levels can be tested


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13711 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-08 21:33:33 +00:00
giannozz 8e5df34485 Calculation of kernel file for tests wasn't working as expected due to dumb 
errors in the script; test "vdw6" works again


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13709 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-08 12:44:22 +00:00
sponce 7f83632575 Disable vdw6 test for now since the 
rVV10_kernel_table seems to be missing. 



git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@13707 c92efa57-630b-4861-b058-cf58834340f0
 2017-08-08 11:02:30 +00:00