| commit | 30c6a00a0383d970ae7dbbcbfba72e8f8700918c | |
| author | Berk Hess <hess@kth.se> | |
| Tue, 9 May 2017 20:45:18 +0000 (9 22:45 +0200) | ||
| committer | Erik Lindahl <erik.lindahl@gmail.com> | |
| Sun, 14 May 2017 14:11:53 +0000 (14 16:11 +0200) | ||
| tree | 55d3bd7240d1907688e5e772280d5bbe8089f075 | treesnapshot (tar.gz zip) |
| parent | 12873e00ffa3c14529e029e7fd199f37f48f3a56 | commitdiff |
Fix bugs in double Simd4 implementations
The double precision version of reduce() and dotProduct()
returned a float with AVX_128_FMA, AVX_512, MIC and IBM_QPX.
Only reduce() is used in double, in the PME force gather.
The same bugs were fixed before for AVX(2)_256.
Refs #2162.
Change-Id: If1a5f044fc70ba1e8d7928a61e4e2c3d170a3922
The double precision version of reduce() and dotProduct()
returned a float with AVX_128_FMA, AVX_512, MIC and IBM_QPX.
Only reduce() is used in double, in the PME force gather.
The same bugs were fixed before for AVX(2)_256.
Refs #2162.
Change-Id: If1a5f044fc70ba1e8d7928a61e4e2c3d170a3922