Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | EIGEN_STRONG_INLINE was NOT inlining in some critical needed areas (6.6X ↵ | Chip-Kerchner | 2021-06-16 |
| | | | | slowdown) when used with Tensorflow. Changing to EIGEN_ALWAYS_INLINE where appropiate. | ||
* | Fix address of temporary object errors in clang11. | Chip Kerchner | 2021-04-02 |
| | | | | This fixes the problem with taking the address of temporary objects which clang11 treats as errors. | ||
* | Fixed performance issues for complex VSX and P10 MMA in gebp_kernel (level 3). | Chip Kerchner | 2021-03-25 |
| | |||
* | Having forward template function declarations in a P10 file causes bad code ↵ | Chip-Kerchner | 2021-02-24 |
| | | | | in certain situations. | ||
* | Add support for dynamic dispatch of MMA instructions for POWER 10 | Pedro Caldeira | 2020-11-12 |