aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/compiler/xla/service/gpu/gpu_layout_assignment.cc
Commit message (Expand)AuthorAge
* Add custom call with layout constraints.Gravatar Mark Heffernan2018-10-08
* [XLA:GPU] Remove hidden flag for disabling heuristic layout assignment.Gravatar Justin Lebar2018-10-05
* Add cuDNN fused convolution forward support.Gravatar Tim Shen2018-09-24
* Further simplify the cuDNN wrappers. Instead of passing aroundGravatar Tim Shen2018-09-24
* [XLA:GPU] Forward batched dot to cublas instead of expanding itGravatar Benjamin Kramer2018-08-03
* [XLA:GPU] Move IsVoltaOrLater into stream_executor_util.h.Gravatar Justin Lebar2018-07-30
* [XLA] Use se:: rather than stream_executor:: in a few places.Gravatar Justin Lebar2018-07-27
* Always use either kAllNCHW or kAllNHWC for GPU convolution layout assignment.Gravatar A. Unique TensorFlower2018-07-27
* [XLA:GPU] Clarify HeuristicLayoutAssignment function.Gravatar Justin Lebar2018-07-25
* Fix a typo in comment to mention kOutputInputYX means NCHWGravatar Smit Hinsu2018-06-29
* [XLA:GPU] Propagate layouts in a better order for performance and fusion.Gravatar Blake Hechtman2018-06-04
* Add heuristic on picking NHWC layout for (V100, fp16) convolutions.Gravatar A. Unique TensorFlower2018-05-24
* Internal changeGravatar Justin Lebar2018-02-02
* [XLA:GPU] Implement BatchNormThunk as a call into cudnn.Gravatar Justin Lebar2017-12-21
* [XLA:CPU] Rename cpu/layout_assignment to cpu/cpu_layout_assignmentGravatar Sanjoy Das2017-12-07
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-01 08:46:41 +0000