Diffusion Lenstool-HPC (chi_GPU) History

R1448/GradientBenchmark/gradient.cpp
chi_GPUc99f2e455a51e84182e1bdc…

Nov 7 2016

removed small control cout 2
e57bf8416890
·schaefer on Nov 7 2016, 16:22.
removed small control cout
f6596bd9e398
·schaefer on Nov 7 2016, 16:22.
added nw test and corrected AVX bug (code tool only multiple of 4 into account
cd34a2adf1b5
·schaefer on Nov 7 2016, 16:18.
adapted tests for modifications
1d3a53f30e9e
·schaefer on Nov 7 2016, 15:00.
merge gilles branch
160bc00c215c
·schaefer on Nov 7 2016, 14:41.

Nov 4 2016

modified the code so that it runs on regor too
ee91f6c1daf3
·branch: gilles
·schaefer on Nov 4 2016, 14:50.
adding vector operators for the intel compilers to ensure backward compatibility
c2f3e43b96e0
·fourestey on Nov 4 2016, 10:52.

Nov 2 2016

added hand optimized inverse and sqrt operations + cleanup
91114be53829
·fourestey on Nov 2 2016, 14:13.

Oct 31 2016

vectorization update with RCP + newton raphson methode to increase the precision
182455bb75cd
·fourestey on Oct 31 2016, 11:06.

Oct 28 2016

hand vectorized version added: 50% faster than the scalar one probably because…
871038e9d982
·fourestey on Oct 28 2016, 15:07.

Oct 25 2016

full loop expension
3e406d2d12ed
·fourestey on Oct 25 2016, 16:09.

Oct 24 2016

added inlined first version of the kernel
d1d407485259
·fourestey on Oct 24 2016, 22:18.

Oct 22 2016

sperated main and compute kernels
5c66f7fce251
·fourestey on Oct 22 2016, 22:45.