1887495434
Merge branch 'develop' into fix-mdi-qmmm
2023-03-12 00:53:40 -05:00
c96ac858bf
GPU Package: Adding JIT test for OpenCL prefetch support.
2023-03-07 21:43:19 -08:00
68ed9ec85e
Merge branch 'develop' into collected-small-changes
2023-03-07 19:39:26 -05:00
d7c783560a
GPU Package: Fixing out of bounds memory access issues with special kernel unroll optimizations.
2023-03-06 22:11:10 -08:00
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
f658a947c6
gpu: fix broken HIP code path
2023-03-05 19:35:32 -07:00
646ef15d83
more consistent way to disable fast math for Intel OpenCL with double precision
2023-03-05 07:18:12 -05:00
2aff321187
Revert "use sin(x+pi/2) instead of cos(x) on Intel OpenCL with double precision"
...
This reverts commit fa38047749
2023-03-05 07:04:20 -05:00
14cb8576f4
Update Kokkos library in LAMMPS to v4.0
2023-03-03 09:22:33 -07:00
9cdb9283ca
another workaround for Intel GPU with OpenCL and double precision
2023-03-03 03:08:03 -05:00
fa38047749
use sin(x+pi/2) instead of cos(x) on Intel OpenCL with double precision
2023-03-02 18:47:34 -05:00
2ccfe635ce
Removed the outdated CUDA_PROXY flag, using CUDA_MPS_SUPPORT consistently in CMake and traditional builds
2023-03-01 16:38:50 -06:00
2511872823
Updated GPU.cmake and Makefile.cuda_mps to not error out with exclusive process mode with the CUDA MPS server in use
2023-02-28 00:41:47 -06:00
ea73ddcb59
remove all files and dirs relevant to the fix latte command
2023-02-24 16:13:50 -07:00
f775de1884
Fix Kokkos compile error with NVIDIA Hopper GPU
2023-02-15 10:02:25 -07:00
88ccd546d8
Fixed bugs with gauss/gpu in bonded systems, including factor_lj in forces and energies
2023-02-09 23:55:14 -06:00
04bed1a6e0
roll back changes for vec3_scale() and vec3_scaleadd() and use temporary vector
2023-02-08 20:32:47 -05:00
09099dd29f
correct preprocessor logic for non-Linux machines
2023-02-08 16:45:16 -05:00
f67d378230
update comment and architecture name
2023-02-08 14:45:50 -05:00
912f046cd7
recover compilation of tersoff kernels with CUDA
2023-02-08 11:16:46 -05:00
30abe68c82
recover kernel failure for tersoff with mixed and single precision
2023-02-08 09:13:04 -05:00
3b4c873beb
another OpenCL bugfix attempt from Trung
2023-02-07 17:31:43 -05:00
d170f83c6d
add experimental Ada CUDA architecture support for conventional make builds.
2023-02-07 13:31:55 -05:00
bde2867251
Make Kokkos configuration compatible with RTX40x0 generation consumer GPUs
...
@stanmoore1 is this the correct way add this?
2023-02-07 08:40:04 -05:00
acf7f9184d
fix failing unit tests with OpenCL
2023-02-06 18:40:59 -05:00
0885edc154
re-enable new neighbor lists with CUDA 12.0 and later
2023-02-05 03:02:19 -05:00
12079c32de
Merge branch 'develop' into fix-mdi-qmmm
2023-02-02 13:10:55 -07:00
3d3bd0d7f2
Merge branch 'develop' into collected-small-fixes
2023-02-02 02:40:50 -05:00
b8b5e385b6
Merge pull request #3630 from akohlmey/download-fallback
...
Fallback URLs for downloading external libraries
2023-02-01 14:40:13 -05:00
89b37c51df
implement download fallback for traditional make build
2023-02-01 06:47:25 -05:00
957f98ddb7
Merge branch 'develop' into collected-small-fixes
2023-01-31 20:42:05 -05:00
83b578f604
BUGFIX: update ML_PACE library version (that fix compilation issue with nvcc)
...
extra update doc/src/pair_pace.rst
2023-01-31 21:08:32 +01:00
e7bcdc63e5
add instructions for running 2-water QMMM PySCF problem
2023-01-31 07:43:20 -07:00
daf23068df
update n2p2 lib version for traditional make, too.
2023-01-31 07:40:13 -05:00
4f0245d542
make Kokkos lib compatible with musl-libc
...
Note: this was adapted from https://github.com/kokkos/kokkos/pull/5678
to be usable without requiring C++17
2023-01-27 12:21:13 -05:00
34e8a74989
apply changes suggested by clang-tidy
2023-01-27 11:49:43 -05:00
b7dfa3db05
Whack extra files from #3532
2023-01-26 15:12:34 -07:00
5bd7b95e60
Merge pull request #3613 from lammps/gpu-neigh-macos
...
Attempted to allow GPU acceleration on MacOS with neighbor builds on …
2023-01-25 13:34:52 -05:00
6fefd8821a
Attempted to allow GPU acceleration on MacOS with neighbor builds on the device by enforcing the old neighbor list code path (will revisit)
2023-01-25 10:42:55 -06:00
722e583b59
use available introspection API to get accumulator data type. update name of flag.
2023-01-25 05:22:49 -05:00
adf43d7fee
Fixed the issues with some OpenCL implementation to avoid errors casting changing the pointer address spaces
2023-01-25 00:02:25 -06:00
64b5ad8966
Merge branch 'develop' into amoeba-gpu
2023-01-24 20:22:59 -05:00
aaa918cbe7
Fixed bugs with access mode on the host side of thetai[1-3]
2023-01-24 17:05:48 -06:00
5014e04341
Removed commented out code, ensured that ic_kspace is not nullptr when call precompute_kspace for hippo/gpu
2023-01-24 08:40:08 -06:00
27da716852
getting out of the insurance business :-)
2023-01-23 16:45:41 -05:00
8e79e2efa5
More cleanup, fixed bugs with hippo fphi kernels for mixed precision
2023-01-23 00:18:42 -06:00
658328dd9d
Added a note in the amoeba doc page on the not-yet resolved issue with integrated GPUs, removed commented out and debugging stuffs in the AM/HP kernels
2023-01-22 17:24:15 -06:00
4d545b3539
remove Fortran library and the styles based on it from MESONT package
2023-01-20 18:29:54 -05:00
ef692258b4
Merge pull request #3532 from stanmoore1/kk_occupancy
...
Update Kokkos version in LAMMPS to 3.7.1
2023-01-20 17:52:05 -05:00
6477b19702
Backport kokkos 4dab4e0 from @weinbe2
2023-01-20 11:12:09 -07:00
