diff options
| author |  Benoit Steiner <benoit.steiner.goog@gmail.com> | 2016-05-17 09:24:35 -0700 |
|---|---|---|
| committer | Benoit Steiner <benoit.steiner.goog@gmail.com> | 2016-05-17 09:24:35 -0700 |
| commit | e7e64c327785ceffd9da018ee265d761991f9685 (patch) | |
| tree | df61359ed916c4eb95acc0ab2f13821dde037cee /unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h | |
| parent | 5fa27574dd29fc5753a0b9d39b1fe5c15668e658 (diff) | |
Enable the use of the packet api to evaluate tensor broadcasts. This speed things up quite a bit:
Before"
M_broadcasting/10 500000 3690 27.10 MFlops/s
BM_broadcasting/80 500000 4014 1594.24 MFlops/s
BM_broadcasting/640 100000 14770 27731.35 MFlops/s
BM_broadcasting/4K 5000 632711 39512.48 MFlops/s
After:
BM_broadcasting/10 500000 4287 23.33 MFlops/s
BM_broadcasting/80 500000 4455 1436.41 MFlops/s
BM_broadcasting/640 200000 10195 40173.01 MFlops/s
BM_broadcasting/4K 5000 423746 58997.57 MFlops/s
Diffstat (limited to 'unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h')
| -rw-r--r-- | unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h b/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h index 9ea1c78eb..5d67f69f3 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorBroadcasting.h @@ -106,7 +106,7 @@ struct TensorEvaluator<const TensorBroadcastingOp<Broadcast, ArgType>, Device> static const int PacketSize = internal::unpacket_traits<PacketReturnType>::size; enum { - IsAligned = false, + IsAligned = true, PacketAccess = TensorEvaluator<ArgType, Device>::PacketAccess, Layout = TensorEvaluator<ArgType, Device>::Layout, RawAccess = false |