scnc
/
quantum-espresso
mirror of https://gitlab.com/QEF/q-e.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
13,437 Commits 5 Branches 40 Tags 1.8 GiB
Commit Graph

13437 Commits

Author SHA1 Message Date
anoop chandran cfb87ee875 wrong checkout corrected for removing eigsolv form make.depend 2018-03-06 17:09:08 +01:00
anoop chandran 808218aa57 removed ifdef from LAXlib makefile[?] 2018-03-06 16:09:48 +01:00
anoop chandran 71e6914a25 because of an error in CI 2018-02-28 23:54:35 +01:00
anoop chandran 0d10f37cdb enable all the proposed tests 2018-02-28 23:52:25 +01:00
anoop chandran ec252e7ada corrections in the Makefiles(LAXlib,PW/src) to account for gpu_eignsolver 2018-02-28 16:07:06 +01:00
anoop chandran 71f224c6b8 added gpu eigensolver dep for object files 2018-02-28 15:58:22 +01:00
anoop chandran 9c1bc3536f merge and build test CI 2018-02-28 15:57:50 +01:00
anoop chandran 99d38aa23d If compiling in GPU, append Eigenslover to MODFLAGS 2018-02-28 15:18:39 +01:00
anoop chandran 322e94c781 corrected a minor mistake 2018-02-06 17:18:00 +01:00
anoop chandran 2e534440b1 changes for power systems 2018-02-06 17:16:43 +01:00
anoop chandran 5edc2b1d19 reactivated template copy for eigensolver-gpu 2018-02-06 15:32:03 +01:00
anoop chandran 40acc4af37 eigenslover-makefile template updated with global make.inc variables 2018-02-06 15:31:07 +01:00
anoop chandran bce6138b29 error caused by the old version eigensolver_makefile template fixed 2018-02-06 14:52:50 +01:00
anoop chandran 22abc3e063 added gpu-eigensolver to main makefile 2018-02-06 14:21:06 +01:00
anoop chandran 672e7976d4 missed a subdir in basemod_flags 2018-02-06 14:08:25 +01:00
anoop chandran 22c46567e7 added makfile template for gpu-eigensolver 2018-02-06 14:04:10 +01:00
anoop chandran b0605c7c87 eigensolver_gpu is added to BASEMOD_FLAGS (not included in LD_LIBS) 2018-02-06 13:51:52 +01:00
anoop chandran 39c272019f Added eigensolver external lib for GPU 2018-02-06 13:50:43 +01:00
Pietro Bonfa 457d408ed7 Build serial, run serial. Plus updated AC_PREREQ to 2.64 2018-01-26 17:32:01 +01:00
Pietro Bonfa faaf18af68 Added gpu builds on Galileo 2018-01-26 16:28:17 +01:00
Pietro Bonfa ef429220b7 GPU eigensolver should not be there by default 2018-01-26 12:39:04 +01:00
Pietro Bonfa 6ce348c438 Initial draft for cuda detection in autotools 2018-01-25 18:49:45 +01:00
Paolo Giannozzi 3a776a7ec1 Merge branch 'master' of https://gitlab.com/QEF/q-e 2018-01-24 18:20:32 +01:00
Paolo Giannozzi 329ebb9def More harmonization of gradient routines: Gamma tricks to spare a FFT, 
use internal variable dft%lgamma instead of global variable gamma_only
(in the future they might have a slightly different meaning), obsolete
routines removed. To be done: CP
 2018-01-24 18:18:15 +01:00
giannozz 09b40c5c46 Merge branch 'newautotools' into 'master' 
configure.ac now compatible with autoconf 2.69

See merge request QEF/q-e!7
 2018-01-24 16:49:00 +00:00
Pietro Bonfa fb0dc5b032 configure.ac now compatible with autoconf 2.69 2018-01-24 17:12:19 +01:00
Stefano de Gironcoli 23893921e1 ignore test.out.* and test.err.* intest-suite when listing modified/untracked files 2018-01-24 10:04:08 +01:00
Stefano de Gironcoli 4c0f7a2baf fft_set_nl and fft_set_nlm unified in a single routine. mill is optional output. 2018-01-24 07:31:29 +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
Stefano de Gironcoli f9905964d4 iinternal indexing of fft_set_nl and fft_set_nlm shifted so that it starts fromm zero. no change in output. 2018-01-24 00:11:04 +01:00
Paolo Giannozzi 308d2baf7c Merge branch 'master' of https://gitlab.com/QEF/q-e 2018-01-23 10:48:09 +01:00
Paolo Giannozzi ab045ac8e0 Added GGA case to normal-mode calculation with phcg.x, example12 2018-01-23 10:46:22 +01:00
Pietro Delugas 01d5b731a9 in the final check of variable cell relaxation the starting_pot flag 
is set to 'atomic' so that potinit do not read a charge density
file with wrong G vectors.
 2018-01-23 10:40:56 +01:00
Paolo Giannozzi 878ac93365 More gradient cleanup (to be completed by porting Gamma tricks into gradutils) 2018-01-21 21:43:07 +01:00
Paolo Giannozzi a1132af040 It looks like CP with LDA+U gets confused when there are atomic wavefunctions 
without U with the same l of atomic wavefunctions with a Hubbard U (example:
TiO2 with U on Ti(3d), ok; FeCuO_2 with U on Fe(3d), not ok). Setting Hubbard_l
to -1 for non-Hubbard atomic wavefunctions should do the job. No warranty.
 2018-01-20 10:12:07 +01:00
Paolo Giannozzi 5f4bf372b6 make.depend updated 2018-01-19 21:05:07 +01:00
Paolo Giannozzi 663e3f1d69 Two more routines computing gradients collected into gradutils, duplicated 
versions deleted: "fft_qgradient" and "fft_qgraddot". These are the versions
of fft_gradient_r2r and fft_graddot for complex arrays with exp(iqr) behavior,
fr usage in the phonon code (gradient of Delta rho at finite q). Note that in
phonon, indices of drho_gradient have been reversed as well in vdW-DF routines,
and that the call to dgradcorr has been changed (dfftp is passed as argument)
 2018-01-18 15:08:46 +01:00
Paolo Giannozzi a88e82dd2a "numerical_gradient" in vdW-DF and rVV10 replaced by "fft_gradent_r2r". 
NOTA BENE: the indices of the array containing the gradient for vdW-DF and
rVV10 functionals have been reverted: now the index of components is the first,
the index  of grid points is the last last. Not sure this is the best choice
but it is the choice done (almost) everywhere else in QE
 2018-01-18 09:50:35 +01:00
Paolo Giannozzi 6e01664329 Added simple test for pw2bgw.x, courtesy Georgy Samsonidze 2018-01-17 18:41:30 +01:00
Paolo Giannozzi c6fecbada0 Test on dual > 4 made more robust towards roundoff errors (dual might turn 
out to be > 4 even when it is supposed to be exactly 4)
 2018-01-17 17:31:04 +01:00
Ivan Carnimeo 3d6fb00960 Merge branch 'master' of gitlab.com:QEF/q-e 2018-01-17 11:53:34 +01:00
Ivan Carnimeo 443d7d342e Bugfix: L-ACE for non cubic cells 2018-01-17 11:53:04 +01:00
Paolo Giannozzi 723dc4ef40 Routine "gradrho" moved to gradutils, with new name "fft_gradient_g2r", 
while previous "fft_gradient" becomes "fft_gradient_r2r".
Routine "grad_dot" moved to gradutils, with new name "fft_graddot" and
removal of useless variable in the list of argument.
 2018-01-16 22:13:52 +01:00
Paolo Giannozzi 4b235efd9d My old compiler doesn't like format "I" with no length specification 2018-01-16 21:37:16 +01:00
Paolo Giannozzi fdbe1f8153 Just in case: better to disable phonons with meta-GGA 2018-01-16 18:29:18 +01:00
Paolo Giannozzi a6742d1d9c pw2gbw.x should work again, courtesy Georgy Samsonidze 2018-01-16 18:28:21 +01:00
Davide Ceresoli 7d1e4f3c0d Improvement of wfck2r: 
- output a subset of wfcs
- output in matlab format
 2018-01-16 15:59:31 +01:00
Paolo Giannozzi cc988c0fe2 Functions gradrho and grad_dot made uniform with other similar routined. 2018-01-15 21:41:21 +01:00
Paolo Giannozzi 663943c646 Too long lines shortened 2018-01-15 14:08:05 +01:00
Paolo Giannozzi b7923806d2 Merge branch 'master' of https://gitlab.com/QEF/q-e 2018-01-15 14:01:43 +01:00