Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Fix oversharding bug in parallelFor. | 2018-06-20 | |
| | |||
* | Add a ThreadPoolInterface* getter for ThreadPoolDevice. | 2018-06-02 | |
| | |||
* | Specialize ThreadPoolDevice::enqueueNotification for the case with no args. ↵ | 2017-10-13 | |
| | | | | As an example this reduces binary size of an TensorFlow demo app for Android by about 2.5%. | ||
* | Moved the choice of ThreadPool to unsupported/Eigen/CXX11/ThreadPool | 2016-12-12 | |
| | |||
* | Reduce dispatch overhead in parallelFor by only calling ↵ | 2016-11-14 | |
| | | | | thread_pool.Schedule() for one of the two recursive calls in handleRange. This avoids going through the scedule path to push both recursive calls onto another thread-queue in the binary tree, but instead executes one of them on the main thread. At the leaf level this will still activate a full complement of threads, but will save up to 50% of the overhead in Schedule (random number generation, insertion in queue which includes signaling via atomics). | ||
* | Avoid unecessary object copies | 2016-08-01 | |
| | |||
* | Return -1 from CurrentThreadId when called by thread outside the pool. | 2016-06-23 | |
| | |||
* | Resolve merge. | 2016-06-23 | |
|\ | |||
* | | size_t -> int | 2016-06-03 | |
| | | |||
* | | Add CurrentThreadId and NumThreads methods to Eigen threadpools and ↵ | 2016-06-03 | |
| | | | | | | | | TensorDeviceThreadPool. | ||
| * | Use signed integers more consistently to encode the number of threads to use ↵ | 2016-06-09 | |
|/ | | | | to evaluate a tensor expression. | ||
* | Fixed compilation error in the tensor thread pool | 2016-05-17 | |
| | |||
* | Merged eigen/eigen into default | 2016-05-17 | |
|\ | |||
| * | Turnon the new thread pool by default since it scales much better over ↵ | 2016-05-13 | |
| | | | | | | | | multiple cores. It is still possible to revert to the old thread pool by compiling with the EIGEN_USE_SIMPLE_THREAD_POOL define. | ||
* | | Address comments by bsteiner. | 2016-05-12 | |
| | | |||
* | | Improvements to parallelFor. | 2016-05-12 | |
|/ | | | | Move some scalar functors from TensorFunctors. to Eigen core. | ||
* | Added a new parallelFor api to the thread pool device. | 2016-05-09 | |
| | |||
* | Provide access to the base threadpool classes | 2016-04-21 | |
| | |||
* | Added the ability to switch to the new thread pool with a #define | 2016-04-21 | |
| | |||
* | Added ability to access the cache sizes from the tensor devices | 2016-04-14 | |
| | |||
* | Prepared the migration to the new non blocking thread pool | 2016-04-14 | |
| | |||
* | Made it possible to customize the threadpool | 2016-03-28 | |
| | |||
* | Fixed compilation error | 2016-03-22 | |
| | |||
* | Pulled latest updates from trunk | 2016-03-22 | |
|\ | |||
* | | Use a single Barrier instead of a collection of Notifications to reduce the ↵ | 2016-03-22 | |
| | | | | | | | | thread synchronization overhead | ||
| * | Fixed a couple of typos | 2016-03-22 | |
| | | |||
| * | Avoid using std::vector whenever possible | 2016-03-22 | |
|/ | |||
* | Split TensorDeviceType.h in 3 files to make it more manageable | 2015-11-20 | |