37f22c8627
Misc Improvements to GPU Package
...
- Optimizations for molecular systems
- Improved kernel performance and greater CPU overlap
- Reduced GPU to CPU communications for discrete devices
- Switch classic Intel makefiles to use LLVM-based compilers
- Prefetch optimizations supported for OpenCL
- Optimized data repack for quaternions
2023-03-05 21:03:12 -08:00
8af77c690c
Merge branch 'develop' into amoeba-gpu
2022-12-14 13:16:41 -06:00
b0accb4ebf
replace atoi() with suitable utility functions
2022-12-09 16:56:17 -05:00
e0181e8c46
refactor handling of pair/only flag for GPU and KOKKOS
2022-10-28 16:38:11 -04:00
2132b1d904
update developer reference text
2022-10-24 11:08:26 -04:00
cccf3f6f78
change email for contacting LAMMPS developers, add to homepage checker
2022-10-20 20:36:03 -04:00
2ef6a59c0a
Merge branch 'develop' into amoeba-gpu
2022-10-01 00:38:24 -05:00
67f6adebb7
whitespace fixup with dynamic casts
2022-09-10 13:26:38 -04:00
35c7911808
One minor typo in fix_gpu.cpp; still trying to figure out what's tripping the non-compile elsewhere
2022-08-13 17:49:39 -05:00
a9c68db3ac
Found three remaining typos; I think we should be OK now
2022-08-13 17:31:03 -05:00
29e1595fb4
Updates to BibTeX syntax for anything that needed it that I caught and fixes to address Issue #2597
2022-08-13 04:51:54 -05:00
e66a4b2423
Changed a few; now pulling new ones from commit c2115c2 so I do not replicate
2022-08-13 01:32:49 -05:00
a829d607ce
work around issues with Intel compilers compiling the GPU package
2022-07-16 10:26:56 -04:00
531e553162
Merge branch 'amoeba' into amoeba-gpu
2022-04-22 16:10:24 -05:00
200b4f13c7
use dynamic cast instead of c-style cast when casting from base type to derived class
2022-04-10 18:18:06 -04:00
0931da9cad
Cleaned up comments in fix_gpu.cpp
2021-11-20 08:38:13 -06:00
a06c4767a0
Merge branch 'upstream' into gpu-newton-pair-on
2021-11-20 08:30:39 -06:00
f641b1c659
final chunk of changes to apply utils::logical()
2021-09-23 07:30:40 -04:00
914f035475
a few more converted commands and updates for unit tests
2021-09-21 17:23:41 -04:00
6adac6b637
Merge branch 'master' into parse-logical-keyword
2021-09-20 20:41:48 -04:00
100da95e3a
convert yes/no on/off flags in the package command(s)
2021-09-20 16:15:24 -04:00
b73c9280c9
improve error message
2021-09-20 13:58:48 -04:00
f80df9ae41
a few more (final?) IWYU updates
2021-09-19 09:41:23 -04:00
5bbec337e5
provide more comprehensive suggestions for GPU neighbor list errors
2021-09-16 10:23:44 -04:00
90225153d9
make sure err_flag is initialized
2021-09-16 07:33:34 -04:00
353b3a2bb3
reformat for increased readability
2021-09-16 07:25:04 -04:00
0202b1169a
Minor edits to the error message
2021-09-12 23:08:36 -05:00
745c7089f0
Temporarily commented out the section in the Atom class where FixGPU finds the optimal bin size. This section makes ev_tally4() in Angle different from CPU-only runs, even with a single command "package gpu 1" without any gpu pair style. Need more effort to understand why.
2021-09-03 01:00:29 -05:00
664a07a3fe
disallow GPU neighbor list with hybrid pair styles (which has still problems)
2021-08-29 17:57:30 -04:00
db92844228
Added recent changes to FixGPU to enable newton_pair on
2021-08-25 23:22:23 -05:00
904a2ef910
Reverted the default setting of newton_pair off for FixGPU; newton_pair can be set to be on via command-line options of package gpu
2021-08-19 22:54:06 -05:00
0904ffa813
Enabled newton pair on for gpu pair styles
2021-08-06 01:11:31 -05:00
09c19a936b
update URLs in source code and manual
2021-05-24 14:18:20 -04:00
d1941392d3
disable clang-format for all existing .cpp files (for now)
2021-05-05 23:53:38 -04:00
183b30abd7
use strmatch() consistently to detect respa and verlet runstyles with optional suffix
2021-03-30 18:30:22 -04:00
42c1e264cb
small cleanup in fix gpu
2021-02-25 10:30:26 -05:00
45c782308c
Fixing issue from recent GPU package update with OMP_NUM_THREADS env being overridden in GPU library.
...
Fixing race condition with OpenMP for GPU styles using torque (missed in regression tests due to the first fix)
Documenting GPU package option for setting the number of threads (consistent with USER-INTEL and USER-OMP).
2021-02-18 21:08:18 -08:00
e7e2d2323b
Feb2021 GPU Package Update - GPU Package Files
2021-02-15 08:20:50 -08:00
959f67962d
allow to revert the pair/only setting
2020-12-17 22:59:12 -05:00
96fa85f61c
alternate implementation of pair/only option (for KOKKOS and GPU)
2020-12-17 22:39:36 -05:00
adf74b3a22
use https://lammps.sandia.gov based URLs consistently
2020-10-21 16:45:30 -04:00
0f6a34775c
Merge branch 'master' into symbolic-atom-constants
2020-09-15 17:44:31 -04:00
f1ef7d85a8
T2345: Replace instances of NULL with nullptr
...
The following changes have been applied to src and lib folders:
regex replace: ([^"_])NULL ⇒ \1nullptr (8968 chgs in src, 1153 in lib)
Manually find/change: (void \*) nullptr ⇒ nullptr (1 case)
regex find: ".*?nullptr.*?"
Manually ~14 cases back to "NULL" in src, ~2 in lib
regex finds a few false positive where nullptr appears between two
strings in a function call
2020-09-12 09:34:38 -06:00
9d0c3bc665
replace some numeric constants in Atom and AtomVec classes with enumerators
2020-09-11 19:07:55 -04:00
6113169a47
IWYU based include statement cleanup
2020-09-02 17:39:34 -04:00
96d8d12a88
replace calls to force->*numeric() with the corresponding utils::*numeric()
...
this also removes the now obsolete functions from Force
2020-08-29 20:31:02 -04:00
229203bac0
Supported neighbor list builds for triclinic boxes on the GPU
2020-05-06 08:51:17 -05:00
2d71d9d47b
add citation for GPU accelerated TIP4P style
2020-01-08 16:15:18 -05:00
5b3a15bd94
putting implementation header first in package folder sources
2019-07-03 19:58:25 -04:00
601746b565
restore lost changes to fix gpu from upstream
2019-05-18 13:09:25 -04:00
