aboutsummaryrefslogtreecommitdiffhomepage
path: root/unsupported
Commit message (Collapse)AuthorAge
* Added more testsGravatar Benoit Steiner2016-04-20
|
* Started to implement a portable way to yield.Gravatar Benoit Steiner2016-04-19
|
* Implemented a more portable version of thread local variablesGravatar Benoit Steiner2016-04-19
|
* Fixed a few typosGravatar Benoit Steiner2016-04-19
|
* Fixed a compilation error with nvcc 7.Gravatar Benoit Steiner2016-04-19
|
* Simplified the code that launches cuda kernels.Gravatar Benoit Steiner2016-04-19
|
* Don't take the address of a kernel on CUDA devices that don't support this ↵Gravatar Benoit Steiner2016-04-19
| | | | feature.
* Use numext::ceil instead of std::ceilGravatar Benoit Steiner2016-04-19
|
* Avoid an unnecessary copy of the evaluator.Gravatar Benoit Steiner2016-04-19
|
* Fixed 2 recent regression testsGravatar Benoit Steiner2016-04-19
|
* Use DenseIndex in the MeanReducer to avoid overflows when processing very ↵Gravatar Benoit Steiner2016-04-19
| | | | large tensors.
* Worked around the lack of a rand_r function on windows systemsGravatar Benoit Steiner2016-04-17
|
* Worked around the lack of a rand_r function on windows systemsGravatar Benoit Steiner2016-04-17
|
* Move the evalGemm method into the TensorContractionEvaluatorBase class to ↵Gravatar Benoit Steiner2016-04-15
| | | | make it accessible from both the single and multithreaded contraction evaluators.
* Deleted unnecessary variableGravatar Benoit Steiner2016-04-15
|
* Fixed a few compilation warningsGravatar Benoit Steiner2016-04-15
|
* Merged in rmlarsen/eigen (pull request PR-178)Gravatar Benoit Steiner2016-04-15
|\ | | | | | | Eigen Tensor cost model part 2: Thread scheduling for standard evaluators and reductions.
| * Get rid of void* casting when calling EvalRange::run.Gravatar Rasmus Munk Larsen2016-04-15
| |
* | Fixed compilation errors with msvcGravatar Benoit Steiner2016-04-15
| |
* | Added ability to access the cache sizes from the tensor devicesGravatar Benoit Steiner2016-04-14
| |
* | Added support for exclusive orGravatar Benoit Steiner2016-04-14
| |
| * Eigen Tensor cost model part 2: Thread scheduling for standard evaluators ↵Gravatar Rasmus Munk Larsen2016-04-14
| | | | | | | | and reductions. The cost model is turned off by default.
* | Added missing definition of PacketSize in the gpu evaluator of convolutionGravatar Benoit Steiner2016-04-14
| |
* | Merged in rmlarsen/eigen (pull request PR-177)Gravatar Benoit Steiner2016-04-14
|\| | | | | | | Eigen Tensor cost model part 1.
* | Enabled the new threadpool testsGravatar Benoit Steiner2016-04-14
| |
* | CleanupGravatar Benoit Steiner2016-04-14
| |
* | Prepared the migration to the new non blocking thread poolGravatar Benoit Steiner2016-04-14
| |
| * Improvements to cost model.Gravatar Rasmus Munk Larsen2016-04-14
| |
* | Added tests for the non blocking thread poolGravatar Benoit Steiner2016-04-14
| |
* | Added a more scalable non blocking thread poolGravatar Benoit Steiner2016-04-14
| |
| * Merge upstream updates.Gravatar Rasmus Munk Larsen2016-04-14
| |\ | |/ |/|
| * Eigen cost model part 1. This implements a basic recursive framework to ↵Gravatar Rasmus Munk Larsen2016-04-14
| | | | | | | | estimate the cost of evaluating tensor expressions.
* | Silenced a compilation warningGravatar Benoit Steiner2016-04-14
| |
* | Added tests to validate flooring and ceiling of fp16Gravatar Benoit Steiner2016-04-14
| |
* | Added simple test for numext::sqrt and numext::pow on fp16Gravatar Benoit Steiner2016-04-14
| |
* | Added basic test for trigonometric functions on fp16Gravatar Benoit Steiner2016-04-14
| |
* | Added support for fp16 to the sigmoid functionGravatar Benoit Steiner2016-04-14
| |
* | Made the test msvc friendlyGravatar Benoit Steiner2016-04-14
|/
* Turn a converge check to a warningGravatar Gael Guennebaud2016-04-13
|
* Fixed compilation warnings generated by clangGravatar Benoit Steiner2016-04-12
|
* Fixed the zeta testGravatar Benoit Steiner2016-04-12
|
* Defer the decision to vectorize tensor CUDA code to the meta kernel. This ↵Gravatar Benoit Steiner2016-04-12
| | | | makes it possible to decide to vectorize or not depending on the capability of the target cuda architecture. In particular, this enables us to vectorize the processing of fp16 when running on device of capability >= 5.3
* bug #1197: fix/relax some LM unit testsGravatar Gael Guennebaud2016-04-09
|
* bug #1160: fix and relax some lm unit tests by turning faillures to warningsGravatar Gael Guennebaud2016-04-09
|
* Disabled the use of half2 on cuda devices of compute capability < 5.3Gravatar Benoit Steiner2016-04-08
|
* Created the new EIGEN_TEST_CUDA_CLANG option to compile the CUDA tests using ↵Gravatar Benoit Steiner2016-04-08
| | | | clang instead of nvcc
* Don't test the division by 0 on float16 when compiling with msvc since msvc ↵Gravatar Benoit Steiner2016-04-08
| | | | detects and errors out on divisions by 0.
* Renamed float16 into cxx11_float16 since the test relies on c++11 featuresGravatar Benoit Steiner2016-04-07
|
* Added missing EIGEN_DEVICE_FUNC to the tensor conversion code.Gravatar Benoit Steiner2016-04-07
|
* Worked around numerical noise in the test for the zeta function.Gravatar Benoit Steiner2016-04-07
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-03 21:01:23 +0000