| commit | 3b52c4ab1954f1492f93b3d7993d81c9f31e71b0 | |
| author | Berk Hess <hess@kth.se> | |
| Tue, 13 Oct 2015 13:54:58 +0000 (13 15:54 +0200) | ||
| committer | David van der Spoel <davidvanderspoel@gmail.com> | |
| Thu, 15 Oct 2015 05:36:26 +0000 (15 07:36 +0200) | ||
| tree | 89185d343b91e65b76b2ea821b603494b199c43c | treesnapshot (tar.gz zip) |
| parent | fd7f8e7e8bb25ca6564c4a598970f9b37dd59204 | commitdiff |
Fixed nbnxn_4xN performance regression
Commit 8e92fd67 changed the 2xNN kernel to use gmx_simd_blendnotzero_r
and the 4xN kernel to use gmx_simd_blendv_r. Making the 4xN kernel
consistent with the 2xNN kernel improves the performance with AVX2
with 4% and 3% for the RF and PME kernels, respectively.
Change-Id: Iac334865c2b2340493639300d07e7ab9c78e129f
Commit 8e92fd67 changed the 2xNN kernel to use gmx_simd_blendnotzero_r
and the 4xN kernel to use gmx_simd_blendv_r. Making the 4xN kernel
consistent with the 2xNN kernel improves the performance with AVX2
with 4% and 3% for the RF and PME kernels, respectively.
Change-Id: Iac334865c2b2340493639300d07e7ab9c78e129f
| src/gromacs/mdlib/nbnxn_kernels/simd_4xn/nbnxn_kernel_simd_4xn_inner.h | diffblobblamehistory |