b5daf22c26Added interface function acSynchronizeMesh
jpekkila
2019-08-12 10:25:05 +03:00
8bbb2cd5dfNow prints device info before trying to run the dummy kernel
jpekkila
2019-08-12 09:46:37 +03:00
e7ff447510Merge branch 'master' into sink_20190723
JackHsu
2019-08-12 14:01:19 +08:00
b782ead4f3Now initilize VTXBUFF_ACCRETION to 0.0, and accretion now works as expected.
JackHsu
2019-08-12 12:34:38 +08:00
fdadd463b7Included the user-defined header after the definition of AcReal to make it available if needed.
jpekkila
2019-08-09 17:11:21 +03:00
bc0276ba2dMerged in dsl_syntax_revision_08008 (pull request #5)
jpekkila
2019-08-09 05:16:37 +00:00
a7a7b535d6The code generator now generates more efficient CUDA: built-in parameters are passed as const references without copying
jpekkila
2019-08-08 21:35:07 +03:00
a172abeb1fModified the other sps files to use the new syntax. Though does not compile since there are some old/very old changes in the DSL that have not been updated to these files (f.ex. RK macro does not exist anymore, it's currently rk3)
jpekkila
2019-08-08 21:25:45 +03:00
67358fa75fRemoved acc/samples as unmaintained
jpekkila
2019-08-08 21:22:34 +03:00
b53cabbc44Made the DSL syntax less confusing: Input and output arrays are now ScalarField and VectorFields instead of scalars and vectors. C++ initializers are now also possible, removing the need to declare Fields as int or int3 which was very confusing, like "what, you assing an int value to a real, what the &^%@?"
jpekkila
2019-08-08 21:07:36 +03:00
e5540f6278added some vertex buffer stuff, not completed.
JackHsu
2019-08-08 14:08:42 +08:00
baac7c8d12Merge branch 'master' into sink_20190723
JackHsu
2019-08-08 12:17:28 +08:00
af24f17d1fMerged in node_device_interface_revision_07-23 (pull request #4)
jpekkila
2019-08-08 02:52:31 +00:00
322cdce52cAdded some new comments + some helpful old comments from a time before the interface revision
jpekkila
2019-08-07 20:05:54 +03:00
2b3f9d75afEnsured that acBoundcondStep is called everywhere in the program before acStore
jpekkila
2019-08-07 19:20:40 +03:00
d8eb2723b4Added an acBoundconds() call before acStore in autotest.cc
jpekkila
2019-08-07 19:10:04 +03:00
1525e0603fAdded some preliminary pragma omps and verified that acIntegrate works as it should.
jpekkila
2019-08-07 19:08:52 +03:00
c2bd5ae3e6Simplified the optimized multi-GPU integration function
jpekkila
2019-08-07 18:17:03 +03:00
fd94b6321dRenamed globalGrid.n to globalGridN
jpekkila
2019-08-07 18:16:34 +03:00
10200e4dd5Merge branch 'master' into node_device_interface_revision_07-23
jpekkila
2019-08-07 16:25:33 +03:00
5432d20b1bRemoved a depreceated auto-optimization script. The optimization is done during acInit() at runtime instead
jpekkila
2019-08-07 16:25:03 +03:00
c38218da3bSRUN_COMMAND has to be exported for it to be available in autotest.sh
jpekkila
2019-08-07 16:22:55 +03:00
b61617ee0fEnabled upwinding by default and updated the model helical forcing with the hotfixed changes from earlier commits. Autotests kinda pass (we get 1 failure but this is likely due to inaccuracies of the trigonometric functions used in helical forcing. The error is very close to the acceptable error bound).
jpekkila
2019-08-07 15:53:32 +03:00
0bb568642fStill one bug
Miikka Vaisala
2019-08-07 19:10:39 +08:00
3adec853dfmade changed to accretion_profile function, still not fully finished but close.
JackHsu
2019-08-07 13:59:12 +08:00
a930864f42Merge branch 'master' into node_device_interface_revision_07-23
jpekkila
2019-08-07 07:43:28 +03:00
cf6b75f82aMerged in cmakelist_rewrite_and_C_API_conformity_07-26 (pull request #1)
jpekkila
2019-08-07 06:53:17 +03:00
2a26b58c38Attempted to add accretion_profile function, and made it adjustable within astaroth.conf
JackHsu
2019-08-07 11:21:24 +08:00
9af5ba2156Copied elements in the DSL form.
Miikka Vaisala
2019-08-07 11:11:27 +08:00
e199ca196dMerge branch 'master' into sink_20190723
Miikka Vaisala
2019-08-07 10:46:23 +08:00
a6fca069a7Added a comment about helical forcing
jpekkila
2019-08-06 19:47:03 +03:00
6b53eb31efErrors with forcing now down from 3 to 1 after switching from fast & inaccurate trig functions to more accurate ones
jpekkila
2019-08-06 19:29:40 +03:00
d7e26e8f21Added forcing from stencil_process.sps to autotests. 3 Tests fail.
jpekkila
2019-08-06 19:15:28 +03:00
1e9ac6edf0Added comments to the autotesting script
jpekkila
2019-08-06 17:43:39 +03:00
e4d9898f35Added improvements to autotest.cc
jpekkila
2019-08-06 17:40:27 +03:00
614a0a1198Added an autotesting script which tests for hydro, magnetic and mhd. Currently hydro and magnetic fail but full mhd works. This indicates that the equations in the hydro and magnetic conditionals have been changed but the autotests have not been updated to correspondingly
jpekkila
2019-08-06 17:40:02 +03:00
3726847683Made globalGridN and d_multigpu_offsets built-in parameters. Note the renaming from globalGrid.n to globalGridN.
jpekkila
2019-08-06 16:39:15 +03:00
812b5e170eAdded some error checking to rendering
jpekkila
2019-08-06 16:32:36 +03:00
b2632c87b4Merge branch 'cmakelist_rewrite_and_C_API_conformity_07-26' into node_device_interface_revision_07-23
jpekkila
2019-08-06 15:18:33 +03:00
51794b8fe0Merge branch 'cmakelist_rewrite_and_C_API_conformity_07-26' into node_device_interface_revision_07-23
jpekkila
2019-08-06 15:17:53 +03:00
280804a438Merge branch 'master' into cmakelist_rewrite_and_C_API_conformity_07-26
jpekkila
2019-08-06 15:14:33 +03:00
e4b981fc62Removed the O2 flag since cmake still defines the O3 flag in CMAKE_CXX_FLAGS_RELEASE and it's confusing which one gcc chooses if both O3 and O2 are passed during compilation. If the issue was the march=native flag then this should also work on Tiara
jpekkila
2019-08-06 14:59:41 +03:00
5f4246fb42Standalone now uses O2 optimization level instead of O3. Also removed -march=native since this causes issues if the program is compiled on a different architecture than it is run on. Since we do not do heavy arithmetic on the host side and the host code is not performance-critical part of the code, -march-native is not very useful anyways
jpekkila
2019-08-06 14:46:13 +03:00
aebf4f11aeAdded function for truelove-density, and took out the unecessary use of switch on-off module for accretion.
JackHsu
2019-08-06 16:17:11 +08:00
632819764eSome intuitions about accretion process. Also defined as a module.
JackHsu
2019-08-06 11:47:55 +08:00
4a846d38f7Merge branch 'master' into sink_20190723
Miikka Vaisala
2019-08-06 10:45:00 +08:00
b73c2675e8Added the optimized implementation of acNodeIntegrate where boundconds are done before integration instead of after
jpekkila
2019-08-05 20:10:13 +03:00
da76fca0dcAdded a test for building an MPI project. Building for the MPI and C API tests is now also disabled by default.
jpekkila
2019-08-05 18:26:12 +03:00
66fa346118Added a copyright text to ctest
jpekkila
2019-08-05 18:24:55 +03:00
b21d2040daAdded a test for building an MPI project. Building for the MPI and C API tests is now also disabled by default.
jpekkila
2019-08-05 18:26:12 +03:00
b1176a0c06Added a copyright text to ctest
jpekkila
2019-08-05 18:24:55 +03:00
fa6e1116cbThe interface revision now actually works. The issue was incorrect order of src and dst indices when storing the mesh.
jpekkila
2019-08-05 17:26:05 +03:00
5232d987c1Added acStoreWithOffset to the revised interface
jpekkila
2019-08-05 16:18:22 +03:00
f3de2fa03cMade globalVertexIdx available during preprocessing. NOTE: potentially dangerous. globalVertexIdx should never be used for reading data from the vertex buffers.
jpekkila
2019-08-05 15:03:02 +03:00
5f89b698ffEmptied the notebook output.
JackHsu
2019-08-05 14:41:30 +08:00
0491f83e8bAdded the correct file this time.
JackHsu
2019-08-05 14:38:46 +08:00
fa99b86601Added a stable astaroth.conf template for testing.
JackHsu
2019-08-05 14:26:32 +08:00
ad53a0b5f0sink_particle.md edited online with Bitbucket
Miikka Väisälä
2019-08-05 05:24:59 +00:00
6dfd03664dStill does not work. I'm starting to think that instead of this one huge revision, we should modify the existing interface step-by-step.
jpekkila
2019-08-02 15:31:24 +03:00
5f2378e91bNow compiles (does not work though)
jpekkila
2019-08-02 15:15:18 +03:00
567ad61465Multinode MPI implementation should be done later in its own branch. The focus of this branch is to revise the node and device layers. Commented out references to the Grid layer.
jpekkila
2019-08-02 13:54:54 +03:00
239e8b4d3eCan set the density for intitial condition from astaroth.conf.
JackHsu
2019-08-02 17:31:49 +08:00
jpekkila