diff options
author | A. Unique TensorFlower <gardener@tensorflow.org> | 2018-07-27 10:04:17 -0700 |
---|---|---|
committer | TensorFlower Gardener <gardener@tensorflow.org> | 2018-07-27 10:07:47 -0700 |
commit | 7a7690a6e079bdb39d157f287e3739f51d8bfe93 (patch) | |
tree | 49ed7c8b8fc8569ef971a7f1f73fe94740bb77b4 /third_party/eigen3 | |
parent | 96c5b537c331011bccafd87d3f90c870c777f349 (diff) |
Quick FusedBatchNorm performance regression fix:
When not compiled with "--config=opt", or when compiling with "--config=opt --distinct_host_configuration=false" (to skip host-specific optimizations), the following code incurs casting overhead even when T == U,
y.reshape(rest_by_depth).device(d) = x_shifted.template cast<T>();
The fix: explicitly avoid calling cast<T>() if T == U.
PiperOrigin-RevId: 206332285
Diffstat (limited to 'third_party/eigen3')
0 files changed, 0 insertions, 0 deletions