Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | bug #1424: add numext::abs specialization for unsigned integer types. | 2017-06-09 | |
| | |||
* | Fixed compilation with cuda-clang | 2017-03-06 | |
| | |||
* | Change comparisons to use Scalar instead of RealScalar. | 2016-12-05 | |
| | |||
* | Added support for expm1 in Eigen. | 2016-12-02 | |
| | |||
* | Adding sycl backend for TensorPadding.h; disbaling __unit128 for sycl in ↵ | 2016-12-01 | |
| | | | | TensorIntDiv.h; disabling cashsize for sycl in tensorDeviceDefault.h; adding sycl backend for StrideSliceOP ; removing sycl compiler warning for creating an array of size 0 in CXX11Meta.h; cleaning up the sycl backend code. | ||
* | Added isnan, isfinite and isinf for SYCL device. Plus test for that. | 2016-11-18 | |
| | |||
* | Specialised basic math functions for SYCL device. | 2016-11-17 | |
| | |||
* | Added a specialization of Eigen::numext::real and Eigen::numext::imag for ↵ | 2016-09-22 | |
| | | | | std::complex<T> to be used when compiling a cuda kernel. This is unfortunately necessary to be able to process complex numbers from a CUDA kernel on MacOS. | ||
* | Calls x.imag() instead of imag(x) when x is a complex number since the former | 2016-09-22 | |
| | | | | is a constexpr while the later isn't. This fixes compilation errors triggered by nvcc on Mac. | ||
* | Added several missing EIGEN_DEVICE_FUNC qualifiers | 2016-09-14 | |
| | |||
* | Added the ability to compute the absolute value of a complex number on GPU, ↵ | 2016-09-12 | |
| | | | | as well as a test to catch the problem. | ||
* | Make sure that our log1p implementation is called as a last resort only. | 2016-08-26 | |
| | |||
* | Add overload of numext::log1p for float/double in CUDA | 2016-08-26 | |
| | |||
* | Factorize the 4 copies of tanh implementations, make numext::tanh consistent ↵ | 2016-08-23 | |
| | | | | with array::tanh, enable fast tanh in fast-math mode only. | ||
* | Fix possible overflow and biais in integer random generator | 2016-08-23 | |
| | |||
* | bug #1247: fix regression in compilation of pow(integer,integer), and add ↵ | 2016-06-25 | |
| | | | | respective unit tests. | ||
* | merge PR 194 | 2016-06-23 | |
|\ | |||
* | | Made log1p_impl usable inside a GPU kernel | 2016-06-16 | |
| | | |||
| * | Generalize expr.pow(scalar), pow(expr,scalar) and pow(scalar,expr). | 2016-06-14 | |
| | | | | | | | | Internal: scalar_pow_op (unary) is removed, and scalar_binary_pow_op is renamed scalar_pow_op. | ||
| * | Enable mixing types in numext::pow | 2016-06-10 | |
|/ | |||
* | Cleaned up the fp16 code | 2016-05-24 | |
| | |||
* | Avoid unnecessary float to double conversions | 2016-05-23 | |
| | |||
* | Silenced several double-promotion warnings | 2016-05-22 | |
| | |||
* | Enable and fix -Wdouble-conversion warnings | 2016-05-05 | |
| | |||
* | fpclassify isn't portable enough. In particular, the return values of the ↵ | 2016-04-27 | |
| | | | | function are not available on all the platforms Eigen supportes: remove it from Eigen. | ||
* | Added support for fpclassify in Eigen::Numext | 2016-04-27 | |
| | |||
* | Improved support for trigonometric functions on GPU | 2016-04-13 | |
| | |||
* | Added support for computing cos, sin, tan, and tanh on GPU. | 2016-04-13 | |
| | |||
* | Use numext::abs instead of std::abs in scalar_fuzzy_default_impl to make it ↵ | 2016-04-08 | |
| | | | | usable inside GPU kernels. | ||
* | Fixed the implementation of Eigen::numext::isfinite, Eigen::numext::isnan, ↵ | 2016-04-08 | |
| | | | | andEigen::numext::isinf on CUDA devices | ||
* | Fixed isfinite_impl: NumTraits<T>::highest() and NumTraits<T>::lowest() are ↵ | 2016-04-08 | |
| | | | | finite numbers. | ||
* | Updated the isnan, isinf and isfinite functions to make compatible with cuda ↵ | 2016-04-07 | |
| | | | | devices. | ||
* | Fixed the signature of numext::abs to make it compatible with complex numbers | 2016-04-04 | |
| | |||
* | Added missing cuda template specializations for numext::ceil | 2016-03-29 | |
| | |||
* | Added support for fmod | 2016-03-28 | |
| | |||
* | Change the header guard around certain numext functions to be CUDA specific. | 2016-03-16 | |
| | |||
* | Update MathFunctions/SpecialFunctions with intelligent header guards. | 2016-03-09 | |
| | |||
* | Add certain functions to numext (log, exp, tan) because CUDA doesn't support ↵ | 2016-03-08 | |
| | | | | | | std:: Use these in SpecialFunctions. | ||
* | Made it possible to run the scalar_binary_pow_op functor on GPU | 2016-02-11 | |
| | |||
* | Fix MSVC warning. | 2016-01-28 | |
| | |||
* | bug #1156: fix several function declarations whose arguments were passed by ↵ | 2016-01-27 | |
| | | | | value instead of being passed by reference | ||
* | Add numext::sqrt function to enable custom optimized implementation. | 2016-01-21 | |
| | | | | | | | | This changeset add two specializations for float/double on SSE. Those are mostly usefull with GCC for which std::sqrt add an extra and costly check on the result of _mm_sqrt_*. Clang does not add this burden. In this changeset, only DenseBase::norm() makes use of it. | ||
* | Fix numerous doxygen shortcomings, and workaround some clang -Wdocumentation ↵ | 2016-01-01 | |
| | | | | warnings | ||
* | Use overload instead of template full specialization to please old MSVC | 2015-11-22 | |
| | |||
* | Fix overload instantiation for clang | 2015-11-20 | |
| | |||
* | Workaround MSVC missing overloads of std::fpclassify for integral types | 2015-11-20 | |
| | |||
* | compilation issue | 2015-11-02 | |
| | |||
* | bug #1102: fix multiple definition linking issue | 2015-10-30 | |
| | |||
* | fix copy/paste typo | 2015-10-28 | |
| | |||
* | Enable std::isfinite/nan/inf on MSVC 2013 and newer and clang. Fix isinf for ↵ | 2015-10-28 | |
| | | | | gcc4.4 and older msvc with fast-math. |