Replace at::Half non-vectorized conversions with implementations from FP16 (#14411)
authorChandler Zuo <chandlerzuo@fb.com>
Tue, 4 Dec 2018 22:23:22 +0000 (14:23 -0800)
committerFacebook Github Bot <facebook-github-bot@users.noreply.github.com>
Tue, 4 Dec 2018 22:32:33 +0000 (14:32 -0800)
commit5ed9dfad982a4e20a51fdf30d3620e76b87ecff4
treeb635e4cb48e5a15c81748abfce0cf0c864f853a5
parent2d56df7892ff6588da3fd079ee95d679644d6730
Replace at::Half non-vectorized conversions with implementations from FP16 (#14411)

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14411
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14579

Folded the fp16 codes into c10.

Reviewed By: ezyang

Differential Revision: D13206450

fbshipit-source-id: 472208dd230dc49d33935622ff3286b17eeb0894
c10/Half-inl.h
c10/Half.cpp
c10/Half.h
c10/core/bitcasts.h [new file with mode: 0644]
c10/test/util/Half_test.cpp [new file with mode: 0644]