Antonio Sanchez cc3573ab44 Disable cuda Eigen::half vectorization on host.
All cuda `__half` functions are device-only in CUDA 9, including
conversions. Host-side conversions were added in CUDA 10.
The existing code doesn't build prior to 10.0.

All arithmetic functions are always device-only, so there's
therefore no reason to use vectorization on the host at all.

Modified the code to disable vectorization for `__half` on host,
which required also updating the `TensorReductionGpu` implementation
which previously made assumptions about available packets.
2021-08-31 19:13:12 +00:00
2021-08-24 17:42:39 +00:00
2021-08-03 17:55:03 +00:00
2021-08-25 20:07:48 +00:00
2012-07-15 10:20:59 -04:00

Eigen is a C++ template library for linear algebra: matrices, vectors, numerical solvers, and related algorithms.

For more information go to http://eigen.tuxfamily.org/.

For pull request, bug reports, and feature requests, go to https://gitlab.com/libeigen/eigen.

Description
No description provided
Readme MPL-2.0 116 MiB
Languages
C++ 85%
Fortran 8.6%
C 2.8%
CMake 1.9%
Cuda 1.2%
Other 0.4%