R1448/srcmaster
R1448/src
master
History Graph
History Graph
Commit | Author | Details | Committed | |||
---|---|---|---|---|---|---|
119bb663c7ed | Christoph Schaefer | fitsfiles | Dec 20 2017 | |||
b95dca84f36d | Christoph Schaefer | Factory function for GPU mapping | Dec 20 2017 | |||
baaf7ff134d9 | fourestey | Merge branch 'master' of ssh://c4science.ch/diffusion/1448/lenstool-hpc | Dec 13 2017 | |||
c9055b6bd7eb | fourestey | adding fast sqrt reciprocal for SIS | Dec 13 2017 | |||
8ae0fc9ab5d2 | fourestey | adding Ofast | Dec 13 2017 | |||
aced509b015f | Christoph Schaefer | first draft of mapping function with factory | Dec 13 2017 | |||
dd4bb043a11b | Christoph Schaefer | finished CPU and GPU gradient2 version | Dec 13 2017 | |||
74c36423b38d | fourestey | improved makefiles | Dec 8 2017 | |||
37cd0c654860 | fourestey | bug correction | Dec 8 2017 | |||
42ed32d4273b | fourestey | bug correction | Dec 8 2017 | |||
8812ace8a01e | fourestey | casting literals explicitely to prevent compilers to do it for us (badly) | Dec 8 2017 | |||
2657c2eccf6a | Christoph Schaefer | forgot | Nov 30 2017 | |||
4986e3290c7c | Christoph Schaefer | aesthetic | Nov 28 2017 | |||
a242bf7ca5c5 | Christoph Schaefer | merge lenstool-mpi | Nov 17 2017 | |||
c420784c230e | Christoph Schaefer | merge | Nov 16 2017 | |||
795c7c7dbe57 | Christoph Schaefer | preparing merge2 | Nov 16 2017 | |||
0a4f8f9d3db2 | Christoph Schaefer | preparing merge | Nov 16 2017 | |||
afcd25a2d0ea | Christoph Schaefer | created and finished potfile_SOA style, wololo | Nov 15 2017 | |||
9a77ec943ca9 | Christoph Schaefer | refactored a bit potentialSoA | Nov 15 2017 | |||
d98b18689916 | Christoph Schaefer | finished float refactoring using read in variables | Nov 15 2017 | |||
59238375e15c | Christoph Schaefer | finished potential SOA,started float refactoring using read in variables | Nov 15 2017 | |||
b6f7f1245887 | Christoph Schaefer | renamed everything to hpp because cleaener that way | Nov 13 2017 | |||
8833e539c371 | Christoph Schaefer | refactored read runmode, solved partly weird segfault problem but definitely… | Nov 10 2017 | |||
92977270e8e3 | Christoph Schaefer | Started refactoring readP function, created slave template and on the hunt of a… | Nov 10 2017 | |||
b5f00f85346e | Christoph Schaefer | cleaning up git mistake maybee | Nov 8 2017 | |||
5e3541b7fc2a | Christoph Schaefer | forgot where i am at | Nov 8 2017 | |||
a25eb2dc9699 | Christoph Schaefer | forgot what the point was | Oct 27 2017 | |||
8dd2440cbe83 | Christoph Schaefer | ... | Oct 17 2017 | |||
6c430c0511f0 | fourestey | reworked cleaner interface for gradient and grid gradient computations | Sep 19 2017 | |||
2dd30f9ade72 | fourestey | removing unecessary output | Sep 19 2017 | |||
a3cbba12c572 | fourestey | adding MPI support | Sep 14 2017 | |||
af81888c9def | fourestey | small updates | Sep 14 2017 | |||
d9fcd2dd364f | fourestey | interface update update | Sep 14 2017 | |||
e1f0e469bd87 | fourestey | 1 rcp instead of 2 divisions | Sep 14 2017 | |||
a282c3906a23 | Christoph Schaefer | Clean GPU Model 5 implementation | Sep 1 2017 | |||
63a0f0e0a288 | Christoph Schaefer | Merge remote-tracking branch 'origin/lenstool_mpi' into cleanGPU | Sep 1 2017 | |||
e32259c143e6 | fourestey | deleted default makefile | Sep 1 2017 | |||
74c7b3d026c5 | fourestey | version 5 inserted | Sep 1 2017 | |||
590823430bc0 | Christoph Schaefer | working gradient GPU function table | Aug 31 2017 | |||
7dfbda11fa1e | Christoph Schaefer | Finished float CPU and works, perferct results | Aug 30 2017 | |||
ee0994fcf2a8 | Christoph Schaefer | finished version, weird result | Aug 25 2017 | |||
0e0e0b5e42bd | fourestey | update | Aug 23 2017 | |||
dc6cc7b15333 | Christoph Schaefer | found bug | Aug 14 2017 | |||
7f19b7de35b6 | fourestey | update after the rebase | Aug 8 2017 | |||
8416c26a60c0 | fourestey | debugged makefile | Aug 8 2017 | |||
52d2417b3b1a | fourestey | new makefiles | Aug 8 2017 | |||
80cf6f51f522 | fourestey | updated to reflect the fact that the GPU version of the gradient moved to src | Aug 8 2017 | |||
76fff640d16a | fourestey | adding GPU library | Aug 8 2017 | |||
cd0e99a7f312 | fourestey | a little bit of tabulation to make things clearer | Aug 8 2017 | |||
d993aacd4595 | fourestey | cleanup | Aug 8 2017 | |||
b01eb5ff7652 | fourestey | de-vectorized loop added to the gradient computation | Aug 8 2017 | |||
5d9a50fb8af5 | fourestey | adding already computed rotations into the gradient computation | Aug 8 2017 | |||
52bbec23e4b6 | fourestey | adding already computed rotations into the gradient computation | Aug 8 2017 | |||
7685f14978af | fourestey | added vectorization off toggle flag | Aug 8 2017 | |||
223d0b03be9e | fourestey | adding tabulation for readability | Aug 8 2017 | |||
87d47626fa76 | fourestey | adding rotation using precomputed cosine/sine | Aug 8 2017 | |||
adcb913ab4db | fourestey | bug correction | Aug 8 2017 | |||
e19a37232c4b | fourestey | update before merge | Aug 8 2017 | |||
ce535ac958fc | fourestey | updated avx version to remove overloaded operators | Aug 8 2017 | |||
de25f0af5356 | fourestey | removed cuda includes | Aug 8 2017 | |||
c6d976421b0f | fourestey | updated avx version to remove overloaded operators | Aug 8 2017 | |||
ce180071f842 | fourestey | removed cuda includes | Aug 8 2017 | |||
6f1510a90976 | fourestey | modified debug output | Aug 8 2017 | |||
12a16ee988a0 | fourestey | adding openmp version and rebranding the old version | Aug 8 2017 | |||
393b276d95f7 | fourestey | adding openmp version | Aug 8 2017 | |||
5dc4d685136b | fourestey | adding debug infos (commented for now) | Aug 8 2017 | |||
c8a4b458969b | Christoph Schaefer | resolved merge | Feb 23 2017 | |||
f1e9d2705e47 | Christoph Schaefer | updated Makefile with env variables | Feb 23 2017 | |||
49db4dcbe135 | Christoph Schaefer | greina0 preparation | Feb 23 2017 | |||
9df9e1a1fc10 | schaefer | played auround with chi | Feb 23 2017 | |||
30af87e0951b | schaefer | relieved register pressure on GPUs by precalculating cos sin values | Feb 16 2017 | |||
15fae33691a4 | schaefer | finished cleaning up chi_CPU function, changed lens_SOA structure, implemented… | Jan 31 2017 | |||
db232b68318b | schaefer | unsorted lens function + big benchmark = 219 Img | Jan 31 2017 | |||
53b8a1923fc7 | schaefer | integrating chi_CPU version into master | Jan 30 2017 | |||
a6eebe218f9d | schaefer | integrating chi_CPU version into master | Jan 30 2017 | |||
5b27d5c4aa67 | schaefer | linkage error found, not solved | Jan 13 2017 | |||
7468cb38b897 | schaefer | corrected a bug in gradient( function pointer was of by 1),implemented… | Jan 12 2017 | |||
a710e43288c0 | fourestey | lense type dispatcher enabled for hand coded avx version | Jan 2 2017 | |||
5bdc6523dd48 | fourestey | inserting guard to allow linking with the lenstool library | Jan 1 2017 | |||
300a1434e86d | fourestey | added improved version of lense derivative selection | Dec 30 2016 | |||
59584d508582 | fourestey | added function to call the correct lense type gradient function correctly | Dec 25 2016 | |||
65f98118f131 | fourestey | useful functions | Dec 25 2016 | |||
feb189992011 | fourestey | interface cleanup + SIS implementation for SOA | Dec 22 2016 | |||
43bf390e02a9 | fourestey | updated library + gradient benchmark with lense 81 auto-vectorized version | Dec 21 2016 | |||
bdbea621f3c5 | fourestey | clean repo | Dec 21 2016 | |||
db8a6d8a39c7 | schaefer | bug fix chi | Dec 16 2016 | |||
c4418ad5944f | schaefer | changes so that it works on regor too | Dec 15 2016 | |||
578dba412eb4 | fourestey | bug fix, precision test passed | Dec 13 2016 | |||
ed21c2f7a9d9 | fourestey | updated makefile to create library | Dec 13 2016 | |||
6003e8cd2da6 | fourestey | added sources for the different versions of the derivative | Dec 12 2016 |
c4science · Help