comp-chem/cp2k
1
0
Fork 0
mirror of https://github.com/cp2k/cp2k.git synced 2026-07-27 05:35:28 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions
23789 commits 34 branches 26 tags 445 MiB
Commit graph

80 commits

Author SHA1 Message Date
Ole Schütt
44d4b3fceb Testing: Split LIBTEST directory 2026-03-23 17:48:46 +01:00
Ole Schütt
3d0b1c90e6 Tests: Convert TEST_TYPES to matchers.py 2025-03-01 19:07:38 +01:00
Matthias Krack
f76adaefd7 Indent @-directives and set them upercase 2024-01-16 16:23:54 +01:00
Ole Schütt
2b6ea54d62 Format all input files 2023-12-18 22:41:19 +01:00
Ole Schütt
5a98848444 Remove dependency on single-precision DBCSR matrices 2023-12-01 15:50:30 +01:00
Ole Schütt
aacccca480 DBM: Fix support for zero sized blocks 2023-09-16 00:02:50 +02:00
Ole Schütt
220241f295 DBM: Add LIBTEST for dbm_multiply 2022-06-26 22:18:05 +02:00
Ole Schütt
424cee84db grid: Move tests from LIBTEST into QS/regtest-grid 2021-03-13 16:13:20 +01:00
Ole Schütt
6a330cddb3 Remove tabs and trailing spaces 2021-03-11 20:06:46 +01:00
Matthias Krack
14984e97f2 Accelerate test 2021-03-03 17:16:22 +01:00
Matthias Krack
065e2b7d8d Add another unit conversion test 2021-03-03 17:16:22 +01:00
Matthias Krack
1ca3b7d46f Add another LIBVORI test input 2021-03-03 17:16:22 +01:00
Ole Schütt
b3f54e074c Adjust 2 thresholds in LIBTEST 2020-12-28 20:19:39 +01:00
Martin Brehm
909fc0d2a6 tests: add sanity tests for libbqb and libvori 
since both libraries have self-checks which can be enabled, they act as
sanity checks despite that we don't check explicitly for any values

Co-authored-by: Tiziano Müller <tiziano.mueller@chem.uzh.ch>
 2020-12-08 17:19:36 +01:00
Ole Schütt
b3beb5c220 Adjust 34 thresholds due to faster collocate general kernel 2020-09-23 23:55:27 +02:00
Ole Schütt
a4726aaf59 grid: Support multiple backends and print stats 2020-08-27 16:14:16 +02:00
Ole Schütt
a42c4d12de do_regtest: Make reference values mandatory 2020-01-28 12:07:44 +01:00
Tiziano Müller
b8d84a3028 adapt to OO-style RNG type and drop unnecessary use of pointers 
Fortran does pass-by-reference, but `POINTER` was needed for the
procedures acting on the `rng_stream_type`. With those refactored as
type-bound procedures, almost all uses of pointers can be avoided.

One of the two remaining cases (in `helium_methods.F`) could likely be
avoided as well by using an allocatable and `move_alloc`, but this would
need another nested type like `rng_stream_p_type` but with an
allocatable member instead of a pointer (or a complete redesign of the
RNG setup in that part).
 2020-01-08 11:30:37 +01:00
Patrick Seewald
93c0d982d1
eri_mme: forgot to register new tests in TEST_FILES (#492) 2019-07-26 14:39:53 +02:00
Patrick Seewald
469a92f17a MME integrals: implement new potentials 2019-05-23 17:29:30 +02:00
Joost VandeVondele
a4b471c861 format tuning and regtest 
svn-origin-rev: 17470
 2016-10-09 10:32:01 +00:00
Dorothea Golze
ecf26ec748 SHG integrals: (a|(r-Ra)^(2m)|b) in libtest 
svn-origin-rev: 17249
 2016-08-14 17:57:04 +00:00
Patrick Seewald
e78a29d054 reduce execution time of minimax regtest and make output more verbose 
svn-origin-rev: 17222
 2016-08-11 08:30:48 +00:00
Dorothea Golze
889ab9ec78 SHG integrals: speed-up of [aba] and [abb] integrals + inclusion in libtest 
svn-origin-rev: 17197
 2016-08-05 21:04:08 +00:00
Patrick Seewald
d0e9d9dedc ERI MME: 3-center ERIs ready, partial RI MP2 / RPA interface 
Correctness of method tested against current GPW.
Not yet implemented for general (non-orthorhombic) cells.
Calculation of B matrix in RI MP2 / RPA not yet supported because RI basis not yet transformed with inverse of L.


svn-origin-rev: 17152
 2016-07-23 16:35:42 +00:00
Patrick Seewald
9b98828e0f ERI MME: refactoring towards eri_mme package 
- reducing dependencies
- some preparations for 3-center ERIs
- minor cleanups and optimizations


svn-origin-rev: 17150
 2016-07-23 11:56:25 +00:00
Dorothea Golze
de3a7412e3 SHG integrals: performance improvement, new operators 
svn-origin-rev: 16962
 2016-05-30 10:50:45 +00:00
Patrick Seewald
3dccb97eb5 Extending minimax approximations of 1/x to more than 15 terms in order to enable higher accuracy for larger ranges of x. Minimax module now covered by regtest. Small changes to original minimax module (function for minimax error, partial replacement of ierr variable by asserts). 
svn-origin-rev: 16583
 2016-02-04 18:18:54 +00:00
Patrick Seewald
83cfd24208 Fast method for periodic 2-center electron repulsion integrals. Application in image charge augmented QM/MM (speeding up calculation of image matrix). 
svn-origin-rev: 16178
 2015-10-19 12:14:42 +00:00
Ole Schütt
f0ea28cbd0 Add more fixed reference values to regtests 
svn-origin-rev: 15843
 2015-09-07 11:30:55 +00:00
Joost VandeVondele
988142f000 bug fix: make sure iseed passed to larnv routines respects the needed constraints. 
svn-origin-rev: 15797
 2015-09-01 18:43:15 +00:00
Ole Schütt
bb17fff6cb prettify TEST_FILES files 
svn-origin-rev: 15777
 2015-08-28 11:26:53 +00:00
Alfio Lazzaro
9f9bab5102 Cleanup of the RMA algorithm. Now the subcommunicators are used by default in DBCSR. Added a new keyword use_mpi_filtering in the DBCSR section to switch on/off the filtering in RMA algo 
svn-origin-rev: 15543
 2015-07-01 22:10:08 +00:00
Alfio Lazzaro
d7d0e4f18b DBCSR: completely remove the test dbcsr_ctype.inp. 
svn-origin-rev: 15126
 2015-03-05 12:19:42 +00:00
Alfio Lazzaro
fb4d004590 DBCSR: remove support MPI subtypes (more cleanup). Remove the corresponding test 
svn-origin-rev: 15125
 2015-03-05 12:06:22 +00:00
Joost VandeVondele
ec680bc078 revised way to make random matrix, O(N) for large sparse matrices. 
svn-origin-rev: 15036
 2015-02-06 13:19:17 +00:00
Joost VandeVondele
1af4925a43 improved filling of the random matrix in the test routines 
svn-origin-rev: 14761
 2014-11-26 09:29:07 +00:00
Joost VandeVondele
2377105e4a add a libtest for dbcsr_binary_read/write 
svn-origin-rev: 14759
 2014-11-25 21:08:43 +00:00
Iain Bethune
03c8b12928 Fix regtest name 
svn-origin-rev: 14658
 2014-11-14 22:37:29 +00:00
Iain Bethune
b889e0c0d0 Improved regtest coverage of fftw3 code 
svn-origin-rev: 14617
 2014-11-12 14:20:36 +00:00
Joost VandeVondele
7014a1c8e0 add a few more libtests for dbcsr 
svn-origin-rev: 13851
 2014-05-09 08:29:45 +00:00
Joost VandeVondele
4a63ebbdd6 more kernels (Peter) and additional dbcsr libtests. 
svn-origin-rev: 13305
 2013-11-10 08:01:48 +00:00
Ole Schütt
43f01cab66 DBCSR: Use cuda priority streams to improve overlapping of MPI-communication with GPU-computation 
svn-origin-rev: 13294
 2013-11-08 12:19:09 +00:00
Ole Schütt
94fde4cc5a DBCSR: Rename keyword n_stack_buffers to cuda/n_buffers 
svn-origin-rev: 13266
 2013-11-01 14:37:50 +00:00
Joost VandeVondele
47c6c08af4 regtest input 
svn-origin-rev: 13099
 2013-08-27 07:21:03 +00:00
Joost VandeVondele
886784cf09 add a test for all dbcsr data types 
svn-origin-rev: 13097
 2013-08-26 15:20:55 +00:00
Joost VandeVondele
235b7094af refactoring of the lower level of the DBCSR multiply (Ole Schuett). 
svn-origin-rev: 12737
 2013-03-04 11:24:50 +00:00
Matthias Krack
7c8e988d96 Remove cvs based tidy scripts. Similar svn based script clean_cwd.sh added (see cp2k/tools/). 
svn-origin-rev: 12089
 2011-12-12 15:23:26 +00:00
Teodoro Laino
75d9a91c84 let's try to be consistent whenever possible: 2 arguments needed! 
svn-origin-rev: 11854
 2011-10-17 04:39:31 +00:00
Urban Borštnik
d2993ca2bb DBCSR/CUDA work. 
Allows multiple GPUs per process; Fix for multiple stack binning-simplifies input; CUDA optimizations; DBCSR configuration output after setup;


svn-origin-rev: 11794
 2011-09-26 13:19:07 +00:00