jpekkila
|
1e6740f999
|
Added the equations for hydro only for both CPU and GPU. NOTE: NOT RIGOROUSLY CHECKED FOR CORRECTNESS. I just took the equations used with entropy and removed the terms which included entropy and magnetic fields
|
2019-07-01 18:56:13 +03:00 |
|
jpekkila
|
21e9fc943b
|
Removed all defines from the stencil assembly and processing files and moved them to the new stencil_defines header. There were spaces after some linebreaks which were removed when I autoformatted the code and that's why there seems to be more changes that there actually is.
|
2019-07-01 18:40:13 +03:00 |
|
jpekkila
|
b8869bb848
|
Linked the defines used by the DSL and astaroth.h. Now f.ex. LENTROPY has to be set only once in acc/mhd_solver/stencil_defines.h and it is propagated throughout the whole project. The name for stencil_defines.h is not final and may change if I come up with a better name. In any case, there is now a standard header associated with each DSL solver where the user can set defines, and declare arrays (VTXBUF_LNRHO) and device constants (uniforms).
|
2019-07-01 18:37:56 +03:00 |
|
jpekkila
|
d9be66f65f
|
NOTE: Renamed LINDUCTION to LMAGNETIC throughout the project.
|
2019-07-01 13:40:24 +03:00 |
|
jpekkila
|
ffe9df1bb0
|
LFORCING was turned off in stencil_process and the autotests failed, re-enabled
|
2019-06-26 14:08:56 +03:00 |
|
Miikka Vaisala
|
a574d6e4c3
|
To be friendly with the autotest.
|
2019-06-25 14:50:43 +08:00 |
|
Miikka Vaisala
|
b1a2d9603c
|
Cleaning the forcing part.
|
2019-06-25 14:48:47 +08:00 |
|
Miikka Vaisala
|
44f26cde42
|
Corrected a stencil bug from der6
|
2019-06-25 11:25:14 +08:00 |
|
Miikka Vaisala
|
59043e586b
|
Added a missing switch to stencil_assembly.sas.
|
2019-06-24 18:03:24 +08:00 |
|
Miikka Vaisala
|
55ecdd0ac8
|
Working upwinding!
It works and it looks so much better than without.
|
2019-06-24 16:58:06 +08:00 |
|
Miikka Vaisala
|
d0b27a0347
|
Upwinding terms now compile.
Not tested yet.
|
2019-06-24 16:32:21 +08:00 |
|
Miikka Vaisala
|
57881e5961
|
Implementetion for hyperdiffusion to be tested later.
|
2019-06-21 17:29:06 +08:00 |
|
Miikka Vaisala
|
f2a33bf4f8
|
Starting with derivatives for upwinding.
|
2019-06-21 16:53:04 +08:00 |
|
jpekkila
|
a7515fbbd7
|
Made the simple forcing to scale with dt s.t. it does not explode so easily
|
2019-06-19 16:34:23 +03:00 |
|
jpekkila
|
6da061b854
|
Fixed a bug in the Astaroth compiler: the boilerplate code for kernels was generated for all nested compound statements instead of just the top-level one
|
2019-06-19 16:10:29 +03:00 |
|
jpekkila
|
e580f6f5d7
|
Improvements to GPU forcing (now applied only at substep 2)
|
2019-06-19 16:08:44 +03:00 |
|
jpekkila
|
b5753f0bec
|
Played around with forcing and cleaned the example implementation a bit
|
2019-06-18 17:40:27 +03:00 |
|
jpekkila
|
8864266042
|
Autoformatted all CUDA/C/C++ code
|
2019-06-18 16:42:56 +03:00 |
|
jpekkila
|
123fc78705
|
Disable forcing by default s.t. the autotests pass
|
2019-06-18 16:39:07 +03:00 |
|
jpekkila
|
4ca4dbefdf
|
Added the machinery for implementing forcing with the DSL on multiple GPUs and a simple model solution
|
2019-06-18 16:13:32 +03:00 |
|
jpekkila
|
0e48766a68
|
Added Astaroth 2.0
|
2019-06-14 14:19:07 +03:00 |
|