aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/stream_executor
Commit message (Expand)AuthorAge
...
* [XLA] FP16 Dot support for the CPU and GPU backends.Gravatar Bixia Zheng2018-02-28
* Do not set cudnn batch norm persistent mode when doing inference.Gravatar Yangzihao Wang2018-02-21
* Merge changes from github.Gravatar Ankur Taly2018-02-16
* Add env-var to specify whether to use CUDNN_BATCHNORM_SPATIAL_PERSISTENT for ...Gravatar Yangzihao Wang2018-02-14
* Remove header dependence on cuda_config.h to fix opensource custom op support.Gravatar Gunhan Gulsoy2018-02-09
* Remove cudnn_type parameter from a few template member functions in classGravatar Bixia Zheng2018-02-08
* Merge changes from github.Gravatar Michael Case2018-02-07
* Performance improvements to some GPU code to use shared locks instead of uniq...Gravatar Rohan Jain2018-01-26
* Merge changes from github.Gravatar Sourabh Bajaj2018-01-17
* Avoid unloading kernels that haven't been loaded and fix replay_computation toGravatar A. Unique TensorFlower2018-01-15
* [XLA:GPU] Warn if ptxas or the driver JIT has known bugs.Gravatar Justin Lebar2018-01-08
* Merge changes from github.Gravatar Patrick Nguyen2017-12-28
* Merge changes from github.Gravatar A. Unique TensorFlower2017-12-22
* Adds support in stream executor interface to update the scratch allocator use...Gravatar A. Unique TensorFlower2017-12-22
* Fix padding for int8 fused convolution.Gravatar Jingyue Wu2017-12-21
* [StreamExecutor] Change "variance" to "inv_var" in BatchNormalizationBackward.Gravatar Justin Lebar2017-12-18
* [StreamExecutor] Allow null batch_mean/batch_var in calls to BatchNormalizati...Gravatar Justin Lebar2017-12-18
* Automated g4 rollback of changelist 179260538Gravatar Dandelion Man?2017-12-15
* Automated g4 rollback of changelist 179258973Gravatar A. Unique TensorFlower2017-12-15
* Merge changes from github.Gravatar Dandelion Man?2017-12-15
* Remove Stream::BlockHostUntilDoneWithStatus; all callers use BlockHostUntilDone.Gravatar A. Unique TensorFlower2017-12-15
* Rename StreamExecutorInterface::BlockHostUntilDoneWithStatus to BlockHostUnti...Gravatar A. Unique TensorFlower2017-12-15
* Rename Stream::BlockHostUntilDoneWithStatus to BlockHostUntilDone.Gravatar A. Unique TensorFlower2017-12-13
* Update Stream::BlockHostUntilDone examples and documentation.Gravatar A. Unique TensorFlower2017-12-13
* Stream::BlockHostUntilDone now returns Status rather than bool.Gravatar A. Unique TensorFlower2017-12-13
* Use BlockHostUntilDoneWithStatus in various places.Gravatar A. Unique TensorFlower2017-12-11
* Fix mismatched argument comments to match parameter namesGravatar A. Unique TensorFlower2017-12-11
* Replace StreamExecutorInterface::BlockHostUntilDone with BlockHostUntilDoneWi...Gravatar A. Unique TensorFlower2017-12-09
* Change TraceListener::BlockHostUntilDoneComplete to pass Status* rather than ...Gravatar A. Unique TensorFlower2017-12-07
* Add BlockHostUntilDoneWithStatus, which returns Status rather than bool.Gravatar A. Unique TensorFlower2017-12-06
* [StreamExecutor] When a kernel launch fails, print the kernel's name.Gravatar Justin Lebar2017-12-05
* [StreamExecutor] Add UnqueryableDeviceParams for all nvidia GPUs.Gravatar Justin Lebar2017-12-04
* Delete trailing whitespaceGravatar A. Unique TensorFlower2017-11-27
* Minor cleanup: remove unnecessary GetCudaContext.Gravatar A. Unique TensorFlower2017-11-22
* Merge changes from github.Gravatar Yifei Feng2017-11-22
* Automated g4 rollback of changelist 176615107Gravatar Yifei Feng2017-11-22
* Automated g4 rollback of changelist 176615737Gravatar Yifei Feng2017-11-22
* Remove duplicate propagate_nans_(false).Gravatar Yifei Feng2017-11-22
* Merged commit includes the following changes:Gravatar A. Unique TensorFlower2017-11-22
* Merge changes from github.Gravatar Yifei Feng2017-11-21
* [SE] Delete deprecated MachineManager.Gravatar Chris Leary2017-11-12
* Include memory size info in error message.Gravatar James Qin2017-11-10
* [StreamExecutor] LOG(ERROR) the driver version when cudnnCreate fails.Gravatar Justin Lebar2017-11-10
* Clean up kernels cached by CUDAExecutor.Gravatar Artem Belevich2017-11-07
* Support Cudnn RNN Fp16Gravatar James Qin2017-11-03
* Merge changes from github.Gravatar Andrew Harp2017-11-02
* Prefer cubin over PTX when we launch CUDA kernels.Gravatar Artem Belevich2017-10-31
* Use Windows compatible string comparisons for setting cuda device flags.Gravatar A. Unique TensorFlower2017-10-18
* Add environment variable to enable setting of CUDA context flags.Gravatar A. Unique TensorFlower2017-10-17
* Add an env-var to choose between FP16 and FP32 as the internal compute type f...Gravatar Yangzihao Wang2017-10-10
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-30 06:01:43 +0000