aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/contrib/nccl
Commit message (Expand)AuthorAge
* Move nccl_rewrite.cc back to tf_kernel_library.Gravatar Gunhan Gulsoy2018-09-13
* [TF] Variant improvements.Gravatar Eugene Brevdo2018-09-11
* fix C++ header guards.Gravatar A. Unique TensorFlower2018-08-21
* Make sure tensorrt and nccl kernels compile with newer versions of Eigen.Gravatar A. Unique TensorFlower2018-08-15
* [ROCm] Interface changes for StreamExecutor to support both CUDA and ROCmGravatar Wen-Heng (Jack) Chung2018-07-12
* Merge pull request #20360 from ppwwyyxx/patch-13Gravatar Yifei Feng2018-07-02
|\
* | Fix Windows GPU BuildGravatar A. Unique TensorFlower2018-06-28
| * Fix gradient of nccl_opsGravatar Yuxin Wu2018-06-27
|/
* Load NCCL lib on-demand to facilitate default NCCL version upgrade to 2Gravatar Smit Hinsu2018-06-18
* Deprecate `DeviceBase::GetStepAllocator()` and replace with calls to `GetAllo...Gravatar Derek Murray2018-05-24
* Use tensorflow::se instead of perftools::gputools for StreamExecutor.Gravatar Justin Lebar2018-04-23
* Enabling fp16 for NCCL 1 and 2.Gravatar A. Unique TensorFlower2018-04-09
* Add support for NCCL2. The configure script asks for what version of NCCL to ...Gravatar A. Unique TensorFlower2018-04-05
* Replaced calls to deprecated tensorflow::StringPiece methods with theirGravatar A. Unique TensorFlower2018-04-02
* Remove all_opensource_files. It's not needed any more.Gravatar Martin Wicke2018-03-28
* disbaling timeout in guitarGravatar Olivia Nordquist2018-03-08
* eager: Rename in_eager_mode to executing_eagerly and get rid of in_graph_mode.Gravatar Asim Shankar2018-03-07
* Remove THIRD_PARTY_ from #include guardsGravatar Sanjoy Das2018-01-24
* Fixes nccl_ops_test error introduced in CL 181736230.Gravatar A. Unique TensorFlower2018-01-16
* Moves CUDA-only condition of NCCL further up to dependent targets.Gravatar A. Unique TensorFlower2018-01-12
* Replaces std::random_shuffle by std::shuffle for C++17 compatibility.Gravatar A. Unique TensorFlower2018-01-09
* Rename Stream::BlockHostUntilDoneWithStatus to BlockHostUntilDone.Gravatar A. Unique TensorFlower2017-12-13
* Use BlockHostUntilDoneWithStatus in various places.Gravatar A. Unique TensorFlower2017-12-11
* Make NCCL code ready for NVIDIA's NCCL 2.Gravatar A. Unique TensorFlower2017-11-29
* Use LINKER_INITIALIZED for mutexes with static storage class.Gravatar A. Unique TensorFlower2017-11-21
* Internal change.Gravatar Anna R2017-11-10
* Check GPU availability after creating test session.Gravatar A. Unique TensorFlower2017-11-10
* BUILD cleanup in contrib/...Gravatar A. Unique TensorFlower2017-10-30
* Fix NCCL rewrite bug when rerunning sessions (assigned device id is not stable).Gravatar A. Unique TensorFlower2017-10-24
* Fix breakages in OSS buildsGravatar Shanqing Cai2017-10-23
* Replace NcclReduce/Broadcast ops during graph optimization so that we can gen...Gravatar A. Unique TensorFlower2017-10-15
* Update to NCCL version 1.3.5. Remove temporary buffer for ncclReduce, it's no...Gravatar A. Unique TensorFlower2017-09-19
* Adding NCCL sum op, register all_sum gradient.Gravatar A. Unique TensorFlower2017-09-12
* Enable contrib.nccl test.Gravatar A. Unique TensorFlower2017-09-05
* Optimize convert_to_tensor.Gravatar A. Unique TensorFlower2017-08-28
* Merge changes from github.Gravatar A. Unique TensorFlower2017-08-15
* Make single-parameter constructors explicitGravatar A. Unique TensorFlower2017-05-31
* Disable flaky timeout test.Gravatar A. Unique TensorFlower2017-05-25
* Internal change.Gravatar A. Unique TensorFlower2017-05-17
* Remove unnecessary copies of value parameters.Gravatar Peter Hawkins2017-05-10
* Internal Change.Gravatar A. Unique TensorFlower2017-04-28
* Add nccl to tf.contrib.Gravatar A. Unique TensorFlower2017-03-31
* Change nccl_manager to use ncclCommInitAll.Gravatar A. Unique TensorFlower2017-02-21
* Merge changes from github.Gravatar Benoit Steiner2017-02-08
* Disable nccl_manager_test since it flakes a lot on Jenkins with errors findingGravatar A. Unique TensorFlower2017-02-07
* Add contrib/nccl for using all-reduce collectives across GPUs of a singleGravatar A. Unique TensorFlower2017-01-24
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-21 22:37:55 +0000