ebfe7f68af
More cleanup
2020-07-05 00:20:28 -05:00
c488e6208d
Similar bug fixes to vashishta/gpu
2020-07-04 23:52:32 -05:00
adacc7f2a4
Similar bug fixes for tersoff/mod/gpu and tersoff/zbl/gpu; some cleanup for sw/gpu
2020-07-04 19:08:55 -05:00
9f733fb305
Removed unused variables
2020-07-04 00:23:34 -05:00
6de020805f
Fixed a bug with tersoff/gpu with building the short neighbor list
2020-07-03 23:55:26 -05:00
66271448ae
Merge pull request #2167 from Vsevak/master
...
Update GPU CMake and Makefile scripts for ROCm v3.5
2020-06-26 13:43:32 -04:00
496bd55dc9
Fix lj/cut/tip4p/long/gpu virial
2020-06-25 17:30:51 +03:00
0ba701acd7
Remove trailing space
2020-06-20 17:25:16 +03:00
723b9e6760
Try to keep compatibility with deprecated ROCm hcc compiler
2020-06-20 15:17:54 +03:00
8eed1b27ff
GPU Makefile cleanup
2020-06-19 00:07:10 +03:00
f75b8e1688
Fix Makefile for GPU ROCm 3.5 backend
2020-06-18 00:33:46 +03:00
622db7053e
Merge remote-tracking branch 'origin/master'
2020-06-15 18:38:40 +03:00
f1aaad2b13
remove trailing whitespace
2020-04-27 09:54:05 -04:00
d53e468ce0
do not mix c++ style and c-style comments
2020-04-26 22:50:24 -04:00
0b6e5be379
cosmetic edit to trigger new CI test runs
2020-04-26 00:25:07 -04:00
1295a9ef46
Fixed bugs with special bonds for ellipsoid gpu pair styles
2020-04-25 10:28:18 -05:00
0608002451
Remove trailing whitespace
2020-04-10 15:10:41 -04:00
19a52dc506
Update lib/gpu/.gitignore
2020-03-30 17:36:22 -04:00
373b57cfdc
Change 'echo -e' to 'printf' for greater compatibility
2020-03-30 17:36:00 -04:00
4f6dd381c3
Merge branch 'master' into gpu_hip
2020-03-30 17:15:55 -04:00
649a8cc01a
Fix typo
2020-03-14 13:57:48 +09:00
5a3a5d86d4
reduce compiler warnings and avoid clash with KOKKOS
2020-03-05 14:27:33 -05:00
3298a86bd9
Reduce redundant copying using UCL_H_Vec::view
2020-02-20 18:32:52 +03:00
d024b3d340
Add shared memory optimization
2020-02-12 01:35:44 +03:00
33fc22b752
Fix echo in Makefile
2020-02-03 01:55:20 +03:00
9e8806bd22
Addition to lib/gpu/README for HIP
2020-01-31 21:26:52 +03:00
66c5fa2abd
Merge 'gpu_hip_port' into master
2020-01-28 20:35:08 +03:00
43a9936241
Fix tagint usage
2020-01-20 03:38:58 +03:00
f69a17dd3a
Reduce unnecessary data exchanges
2020-01-19 02:05:16 +03:00
514271c984
recover compilation with -DLAMMPS_BIGBIG and fix whitespace and formatting issues
2020-01-08 17:57:54 -05:00
18a0cd3ead
Merge pull request #1817 from akohlmey/collected-small-fixes
...
Collected small bugfixes and updates
2020-01-08 17:16:34 -05:00
d3eed9dba4
Merge pull request #1752 from ndtrung81/gpu-mem-leak-init
...
Fixed bugs with kernel (re)compiling when the global device got clear and reinitialized
2020-01-08 16:17:13 -05:00
6e2c3b7ce9
Merge branch 'master' into collected-small-fixes
2020-01-08 16:14:19 -05:00
481c647889
Merge pull request #1776 from Vsevak/tip4p_gpu
...
GPU support for lj/cut/tip4p/long pair style
2020-01-08 15:35:39 -05:00
00dce6b198
Merge branch 'master' into gpu-mem-leak-init
2020-01-08 14:26:03 -05:00
e8a941a0b9
Remove unnecessary data transfer from tip4p/gpu pairstyle init
...
Such data transfer is performed at each timestep, so it does not belong to the initialization
2020-01-08 01:48:47 +03:00
66a473455e
silence warnings from generic OpenCL header files (as seen on Fedora 31)
2020-01-06 17:34:26 -05:00
337b0fd649
Fix 'sametag' transfer
2019-12-22 20:20:12 +03:00
4c3ec145f3
Corrected the wrong use of the stubs path in -rpath
2019-12-17 14:05:39 -06:00
7e8a04d985
More minor update to README
2019-12-12 09:57:49 -06:00
99ba15bf6a
Updated README and added -L$(CUDA_HOME)/lib64/stubs to the Makefile's
2019-12-11 16:29:42 -06:00
a3fca53e97
Separate the computation of newsite into another kernel
...
This eliminates the need for thread fence and makes the calculation stable on GTX1070 (CUDA and OpenCL) and TitanV
2019-12-07 15:54:20 +03:00
4a51e1660f
Add thread fence in kernel to fix Volta indeterminacy
2019-12-07 01:16:39 +03:00
a2f9fa8e78
Separate the computation of hneigh into another kernel
...
Simplify the main GPU kernel and add another kernel 'k_pair_reneigh'. It works good on GTX1070 (Pascal), but still there is a problem with non-deterministic results on Volta.
I reimplement BaseCharge::compute methods in the child class LJ_TIP4PLong to correctly embed a new kernel in the code.
Also commit includes some codestyle fixes.
2019-12-06 21:41:02 +03:00
66a076b819
Delete "float.h" from TIP4P GPU kernel for better OpenCL compatibility
...
According to Khronos OpenCL docs, "The C99 standard headers <...>, float.h, <...> are not available and cannot be included by a program"
2019-11-27 19:49:12 +03:00
4febc7f794
Add copyright and fix style
2019-11-18 02:01:32 +03:00
ca8d1ac2ff
Simplify tip4p GPU memory resize using 'resize_ib'
2019-11-18 02:01:32 +03:00
f803ba5655
Add shfl_xor sum to kernel for ARCH>=300
2019-11-12 21:35:36 +03:00
64bdc59623
Implement GPU pair style lj/cut/tip4p/long/gpu
...
Source code, Makefiles and Install for GPU-accelerated TIP4P pair style.
It is implemented as a part of the standard GPU package.
The style is compatible with the standard lj/cut/tip4p/long.
Also, this commit modifies "atom.h" just to
add a getter for variable 'max_same'.
2019-11-10 02:38:58 +03:00
e119bffcca
Fixed bugs with kernel (re)compiling when the global device got cleared and then reinitialized
2019-10-31 01:15:53 -05:00
