diff options
| author |  Gael Guennebaud <g.gael@free.fr> | 2019-01-30 23:45:12 +0100 |
|---|---|---|
| committer | Gael Guennebaud <g.gael@free.fr> | 2019-01-30 23:45:12 +0100 |
| commit | 7ef879f6bfa465a80109216e6d0b18266ef97321 (patch) | |
| tree | 404916cdc86ed7dad6376093fa8fc9324b47027a /unsupported/Eigen/CXX11/src/Tensor/TensorCustomOp.h | |
| parent | de77bf5d6c4fb63a07a7bf7201b26f435d9b19b5 (diff) | |
GEBP: improves pipelining in the 1pX4 path with FMA.
Prior to this change, a product with a LHS having 8 rows was faster with AVX-only than with AVX+FMA.
With AVX+FMA I measured a speed up of about x1.25 in such cases.
Diffstat (limited to 'unsupported/Eigen/CXX11/src/Tensor/TensorCustomOp.h')
0 files changed, 0 insertions, 0 deletions