1251f61570
Removed a stray acBoundcondStep() in acStore where it definitely shouln't be. Removed code duplication: acBoundcondStep now uses the new acLocalBoundcondStep and acGlobalBoundcondStep functions.
2019-07-09 17:08:18 +03:00
6c7b2dbd8d
Merge branch 'master' into multigpu_optimization_2019-07-05
2019-07-09 16:44:55 +03:00
6ec5e5a2c6
bitbucket-pipelines.yml edited online with Bitbucket
2019-07-09 13:21:16 +00:00
bc6f91f610
bitbucket-pipelines.yml edited online with Bitbucket
2019-07-09 12:53:50 +00:00
fba09d2427
bitbucket-pipelines.yml edited online with Bitbucket
2019-07-09 12:51:36 +00:00
afb17c78d4
bitbucket-pipelines.yml edited online with Bitbucket
2019-07-09 12:48:14 +00:00
314f3c1fcc
bitbucket-pipelines.yml edited online with Bitbucket
2019-07-09 12:31:16 +00:00
5ceb8ddb7e
bitbucket-pipelines.yml edited online with Bitbucket
2019-07-09 12:26:08 +00:00
d56e6f1492
Initial Bitbucket Pipelines configuration
2019-07-09 12:23:15 +00:00
10a98b01a9
Experimental change: now the integration function is automatically optimized during acInit
2019-07-09 14:46:24 +03:00
a086821e7c
Added a function acAutoOptimize to the interface and removed rk3_step_async in kernels.cuh (moved into rkStep)
2019-07-09 14:21:22 +03:00
84d96de42b
Merge branch 'master' into multigpu_optimization_2019-07-05
2019-07-09 13:40:33 +03:00
508d15b578
Switched from math.h to cmath in math_utils.h. The old-school C math functions are bugged/not overloaded properly in GCC < 6.0 when compiling C++.
2019-07-09 13:37:08 +03:00
deebe570da
Merge branch 'master' into multigpu_optimization_2019-07-05
2019-07-08 16:11:24 +03:00
eda2f6543b
Created a new ForcingParams structure and some functions for generating and transferring the forcing parameters to the host/device
2019-07-08 15:43:37 +03:00
f9be905703
Corrected an unit coversion issue from forcing.
...
Now noticing these because of switching to gcc 8.
2019-07-08 16:43:37 +08:00
df1ba6264a
Update to readme.
2019-07-08 11:08:45 +08:00
6ba15c3a7c
props.totalConstMem and props.sharedMemPerBlock cause assembler error
...
while compiling on TIARA gp cluster. Therefore commeted out.
2019-07-08 11:00:12 +08:00
5fdfdeca9e
Multi-GPU optimizations: removed some unnecessary synchronization and divided the calculation of boundary conditions to local and global steps.
2019-07-05 18:21:44 +03:00
f1066a2c11
Added preliminary pragmas for dispatching commands simultaneously to multiple GPUs (commented out)
2019-07-05 17:16:12 +03:00
2092adc0f6
Preparations for multi-GPU optimizations
2019-07-05 15:44:30 +03:00
ce8fe53f91
Moved explanations and comments to the beginning of astaroth.cu. No code changes.
2019-07-05 15:39:52 +03:00
d87eb36f5a
Formatting: brackets around a for loop for consistency
2019-07-05 15:26:19 +03:00
224b91b83a
Added more control for synchronizing streams and halos among the GPUs
2019-07-05 15:17:20 +03:00
332f1a4f40
Reordered some of the functions in astaroth.cu and introduced acExchangeHalos() for synchronizing the part of the grid that is independent from the chosen boundary conditions between subgrids.
2019-07-05 15:01:51 +03:00
c71711ec36
Disabled real-time visualization by default. SDL2 is no longer a dependency when building with the default flags.
2019-07-04 22:30:26 +03:00
ad7a497eef
Added a comment about timestepping and autoformat
2019-07-04 17:25:54 +03:00
d1a93b7d4e
acIntegrateStepWithOffset corrected and confirmed to work on 1-4 GPUs
2019-07-04 16:58:24 +03:00
01437411b6
Comment
2019-07-04 16:39:20 +03:00
91f119e8dd
Deprecated the old implementation of acIntegrateStep. acIntegrateStep now calls acIntegrateStepWithOffset instead of device.cuh functions.
2019-07-04 16:37:55 +03:00
5049dadc1c
Implemented acIntegrateStepWithOffset
2019-07-04 16:31:16 +03:00
a53e0a170d
Overloaded max/min for int3 and removed old comments
2019-07-04 16:24:08 +03:00
e1d545b0eb
Code readability and cleanup (remembered that int3 has + and - operators defined in math_utils.h)
2019-07-04 16:16:49 +03:00
30254d9abb
Removed a redundant and old gridIdxx function which I though I already removed a long time ago.
2019-07-04 16:10:29 +03:00
b3a0b10a86
Removed old comments
2019-07-04 16:02:13 +03:00
24f49ce461
Host code now compiled according to the C++11 standard. The device code has used C++11 for a while now and it's good to use a single standard consistently throughout the project. Old Intel compilers (~14.0) do not seem to support C++11 but the code should still compile since the flag is ignored in that case.
2019-07-04 16:01:02 +03:00
0884c4bf38
Moved the definition of acForcingVec to host_forcing.cc since it depends on user parameters that may not be defined in all projects
2019-07-04 15:28:18 +03:00
698d04c57d
Removed a deprecated code block
2019-07-04 15:08:42 +03:00
317df4c025
Re-enabled the check for implicit float-double-float conversions with gcc
2019-07-04 13:45:09 +03:00
edafe5a563
Made the gcc version check more lenient
2019-07-04 13:43:49 +03:00
b65a0750cc
Updated CUDA version on README
2019-07-04 14:14:24 +08:00
ce54499351
Removed an accidental '\'
2019-07-04 14:11:26 +08:00
2c4d129216
-Werror-all disabled from Intel compiler flags
...
Because produces cryptical messages preventing compilation. The messages do not
tell anything about the problem and are also at odds with some gcc specific
pragma commands. If we can resolve these issues, we can turn it back.
2019-07-04 11:25:04 +08:00
0a030742f9
Added support for calling compile_acc.sh with custom headers
2019-07-03 19:06:39 +03:00
7abb959828
Overhaul to the user-defined parameters done: All logical switches, parameters and vertex buffer handles are now defined in a single header file (the default location is acc/mhd_solver/stencil_defines.h). This header is used when preprocessing the DSL sources and is linked to the include/ directory when calling scripts/compile_acc.sh. astaroth.h is now used for configuring internal stuff only and should not be modified by users
2019-07-03 19:01:16 +03:00
6907d74ea3
Suppressed an unused variable warning for globalVertexIdx
2019-07-03 18:46:17 +03:00
7d6255ba14
Suppressed unused variable warnings in kernels.cuh
2019-07-03 18:12:48 +03:00
556cb77dbb
Removed weird unused functions from model_rk3.cc. I have no idea where these came from. Probably remnants from the time I was pulling all-nighters to finish my thesis.
2019-07-03 18:06:40 +03:00
b4eea4b6b6
Changed #if 0 to #if LFORCING instead to get the code to compile if forcing is used (even though autotesting does not support it yet). Also more autoformatting. Maybe I should disable it or then everyone should start using it to avoid cluttering commits with these superficial changes
2019-07-03 17:49:34 +03:00
609cfaea14
Commenting the unused variables in simulation.cc a few commits back was the wrong move since the variables are needed if LFORCING it set.
2019-07-03 17:46:50 +03:00
jpekkila