Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | bug #1382: move using std::size_t/ptrdiff_t to Eigen's namespace (still ↵ | Gael Guennebaud | 2017-01-23 |
| | | | | better than the global namespace!) | ||
* | Made sure that enabling avx2 instructions enables avx and sse instructions ↵ | Benoit Steiner | 2017-01-19 |
| | | | | as well. | ||
* | Include SSE packet instructions when compiling with avx512 enabled. | Benoit Steiner | 2016-12-19 |
| | |||
* | Fixed merge conflicts | Benoit Steiner | 2016-11-19 |
| | |||
* | Merged eigen/eigen into default | Benoit Steiner | 2016-11-17 |
|\ | |||
* | | Specialised basic math functions for SYCL device. | Luke Iwanski | 2016-11-17 |
| | | |||
| * | Enable the use of AVX512 instruction by default | Benoit Steiner | 2016-11-16 |
|/ | |||
* | Merged eigen/eigen into default | Benoit Steiner | 2016-11-04 |
|\ | |||
* | | Removed the sycl include from Eigen/Core and moved it to ↵ | Mehdi Goli | 2016-11-04 |
| | | | | | | | | Unsupported/Eigen/CXX11/Tensor; added TensorReduction for sycl (full reduction and partial reduction); added TensorReduction test case for sycl (full reduction and partial reduction); fixed the tile size on TensorSyclRun.h based on the device max work group size; | ||
| * | Improved AVX512 configuration | Benoit Steiner | 2016-11-03 |
| | | |||
| * | Improved AVX512 support | Benoit Steiner | 2016-11-03 |
| | | |||
| * | Merged eigen/eigen into default | Benoit Steiner | 2016-11-03 |
| |\ | |/ |/| | |||
* | | Disable vectorization on device only when compiling for sycl | Benoit Steiner | 2016-11-02 |
| | | |||
* | | Merge latest updates from trunk | Benoit Steiner | 2016-10-20 |
|\ \ | |||
| * | | Fixed a few typos in the ternary tensor expressions types | Benoit Steiner | 2016-10-19 |
| | | | |||
* | | | Fixing the typo regarding missing #if needed for proper handling of ↵ | Mehdi Goli | 2016-10-16 |
| | | | | | | | | | | | | exceptions in Eigen/Core. | ||
* | | | Applyiing Benoit's comment to return the missing line back in Eigen/Core | Mehdi Goli | 2016-10-14 |
| | | | |||
* | | | Reducing the code by generalising sycl backend functions/structs. | Mehdi Goli | 2016-10-14 |
| | | | |||
| | * | Enabling AVX512 should also enable AVX2. | Benoit Steiner | 2016-10-06 |
| | | | |||
| | * | Merged latest updates from trunk | Benoit Steiner | 2016-10-05 |
| | |\ | | |/ | |/| | |||
* | | | Pull the latest updates from trunk | Benoit Steiner | 2016-10-05 |
|\| | | |||
| * | | Added support for constand std::complex numbers on GPU | Benoit Steiner | 2016-10-03 |
| | | | |||
| * | | Add CUDA-specific std::complex<T> specializations for scalar_sum_op, ↵ | RJ Ryan | 2016-09-20 |
| | | | | | | | | | | | | scalar_difference_op, scalar_product_op, and scalar_quotient_op. | ||
* | | | Merged with default. | Luke Iwanski | 2016-09-19 |
|\| | | |||
* | | | Partial OpenCL support via SYCL compatible with ComputeCpp CE. | Luke Iwanski | 2016-09-19 |
| | | | |||
| * | | Factorize the 4 copies of tanh implementations, make numext::tanh consistent ↵ | Gael Guennebaud | 2016-08-23 |
|/ / | | | | | | | with array::tanh, enable fast tanh in fast-math mode only. | ||
* | | bug #1232: refactor special functions as a new SpecialFunctions module, ↵ | Gael Guennebaud | 2016-07-08 |
| | | | | | | | | currently in unsupported/. | ||
* | | Add TernaryFunctors and the betainc SpecialFunction. | Eugene Brevdo | 2016-06-02 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TernaryFunctors and their executors allow operations on 3-tuples of inputs. API fully implemented for Arrays and Tensors based on binary functors. Ported the cephes betainc function (regularized incomplete beta integral) to Eigen, with support for CPU and GPU, floats, doubles, and half types. Added unit tests in array.cpp and cxx11_tensor_cuda.cu Collapsed revision * Merged helper methods for betainc across floats and doubles. * Added TensorGlobalFunctions with betainc(). Removed betainc() from TensorBase. * Clean up CwiseTernaryOp checks, change igamma_helper to cephes_helper. * betainc: merge incbcf and incbd into incbeta_cfe. and more cleanup. * Update TernaryOp and SpecialFunctions (betainc) based on review comments. | ||
* | | bug #725: make move ctor/assignment noexcept. | Gael Guennebaud | 2016-06-03 |
| | | |||
* | | Include mmintrin.h to make it possible to use mmx instructions when needed. ↵ | Benoit Steiner | 2016-05-23 |
| | | | | | | | | For example, this will enable the definition of a half packet for the Packet4f type. | ||
* | | Started to vectorize the processing of 16bit floats on CPU. | Benoit Steiner | 2016-05-23 |
| | | |||
| * | Pulled latest updates from upstream | Benoit Steiner | 2016-04-29 |
| |\ | |/ |/| | |||
* | | Don't attempt to leverage the _cvtss_sh and _cvtsh_ss instructions when ↵ | Benoit Steiner | 2016-04-20 |
| | | | | | | | | compiling with clang since it's unclear which versions of clang actually support these instruction. | ||
* | | Made sure all the required header files are included when trying to use fp16 | Benoit Steiner | 2016-04-19 |
| | | |||
* | | Merged eigen/eigen into default | Rasmus Larsen | 2016-04-11 |
|\ \ | |||
| | * | Pull latest updates from upstream | Benoit Steiner | 2016-04-11 |
| | |\ | | |/ | |/| | |||
| * | | Remove all references to MKL in BLAS wrappers. | Gael Guennebaud | 2016-04-11 |
| | | | |||
* | | | Merged eigen/eigen into default | Rasmus Larsen | 2016-04-07 |
|\| | | |||
| * | | Added support for hardware conversion between fp16 and full floats whenever | Benoit Steiner | 2016-04-06 |
| | | | | | | | | | | | | possible. | ||
| * | | Merged kmargar/eigen/tip into default | Konstantinos Margaritis | 2016-04-05 |
| |\ \ | |||
| | * | | enable the other includes as well | Konstantinos Margaritis | 2016-04-05 |
| | | | | |||
* | | | | Merged eigen/eigen into default | Rasmus Larsen | 2016-04-01 |
|\| | | | |||
* | | | | Add matrix condition estimator module that implements the Higham/Hager ↵ | Rasmus Munk Larsen | 2016-04-01 |
| | | | | | | | | | | | | | | | | algorithm from http://www.maths.manchester.ac.uk/~higham/narep/narep135.pdf used in LPACK. Add rcond() methods to FullPivLU and PartialPivLU. | ||
| * | | | Pull math functions from the global namespace only when compiling cuda code ↵ | Benoit Steiner | 2016-03-30 |
|/ / / | | | | | | | | | | with nvcc. When compiling with clang, we want to use the std namespace. | ||
| * / | some primitives ported, but missing intrinsics and crash with asm() are a ↵ | Konstantinos Margaritis | 2016-03-27 |
|/ / | | | | | | | problem | ||
* | | Made half floats usable on hardware that doesn't support them natively. | Benoit Steiner | 2016-03-11 |
| | | |||
* | | Added support for simple coefficient wise tensor expression using half ↵ | Benoit Steiner | 2016-02-19 |
| | | | | | | | | floats on CUDA devices | ||
* | | FP16 on CUDA are only available starting with cuda 7.5. Disable them when ↵ | Benoit Steiner | 2016-02-18 |
| | | | | | | | | using an older version of CUDA | ||
* | | Reverted unintended changes introduced by a bad merge | Benoit Steiner | 2016-02-19 |
| | | |||
* | | Added preliminary support for half floats on CUDA GPU. For now we can simply ↵ | Benoit Steiner | 2016-02-19 |
| | | | | | | | | convert floats into half floats and vice versa |