aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Replaced a couple more uint16_t with unsigned shortGravatar Benoit Steiner2016-03-11
|
* Use unsigned short instead of uint16_t since they're more portableGravatar Benoit Steiner2016-03-11
|
* Made half floats usable on hardware that doesn't support them natively.Gravatar Benoit Steiner2016-03-11
|
* Made the comparison of Eigen::array GPU friendlyGravatar Benoit Steiner2016-03-11
|
* Added a comparison operator for Eigen::arrayGravatar Benoit Steiner2016-03-11
| | | | Alias Eigen::array to std::array when compiling with Visual Studio 2015
* Updated the cxx11_meta test to work on the Eigen::array class when ↵Gravatar Benoit Steiner2016-03-11
| | | | std::array isn't available.
* Fixed the +=, -=, *= and /= operators to return a referenceGravatar Benoit Steiner2016-03-10
|
* Worked around visual studio compilation warnings.Gravatar Benoit Steiner2016-03-09
|
* Fixed a typo.Gravatar Benoit Steiner2016-03-10
|
* Made the list reductions less ambiguous.Gravatar Benoit Steiner2016-03-09
|
* Avoid implicit castGravatar Benoit Steiner2016-03-09
|
* Made sure to include the <random> header file when compiling with visual studioGravatar Benoit Steiner2016-03-09
|
* Avoid unnecessary conversion from 32bit int to 64bit unsigned intGravatar Benoit Steiner2016-03-09
|
* Enable the random number generators when compiling with visual studioGravatar Benoit Steiner2016-03-09
|
* Fixed the integer division code on windowsGravatar Benoit Steiner2016-03-09
|
* Fixed static assertionGravatar Benoit Steiner2016-03-08
|
* Replace std::vector with our own implementation, as using the stl when ↵Gravatar Benoit Steiner2016-03-08
| | | | compiling with nvcc and avx enabled leads to many issues.
* Simplified the full reduction codeGravatar Benoit Steiner2016-03-08
|
* Fixed the tensor generator codeGravatar Benoit Steiner2016-03-08
|
* Fixed the tensor concatenation codeGravatar Benoit Steiner2016-03-08
|
* Fixed the tensor layout swapping codeGravatar Benoit Steiner2016-03-08
|
* Fixed the tensor chipping code.Gravatar Benoit Steiner2016-03-08
|
* Decoupled the packet type definition from the definition of the tensor ops. ↵Gravatar Benoit Steiner2016-03-08
| | | | All the vectorization is now defined in the tensor evaluators. This will make it possible to relialably support devices with different packet types in the same compilation unit.
* Use NumTraits::highest() and NumTraits::lowest() instead of the ↵Gravatar Benoit Steiner2016-03-07
| | | | std::numeric_limits to make the tensor min and max functors more CUDA friendly.
* Added the ability to pad a tensor using a non-zero valueGravatar Benoit Steiner2016-03-07
|
* Fix a couple of typos in the code.Gravatar Benoit Steiner2016-03-07
|
* Added a test to validate the behavior of some of the tensor syntactic sugar.Gravatar Benoit Steiner2016-03-07
|
* Added missing includeGravatar Benoit Steiner2016-03-06
|
* Don't try to compile the uint128 test with compilers that don't support uint127Gravatar Benoit Steiner2016-03-06
|
* Don't warn that msvc 2015 isn't c++11 compliant just because it doesn't ↵Gravatar Benoit Steiner2016-03-06
| | | | claim to be.
* Turn on some of the cxx11 features when compiling with visual studio 2015Gravatar Benoit Steiner2016-03-05
|
* Don't test our 128bit emulation code when compiling with msvcGravatar Benoit Steiner2016-03-05
|
* Avoid using initializer lists in test since not all version of msvc support themGravatar Benoit Steiner2016-03-05
|
* Use EIGEN_PI instead of redefining our own constant PIGravatar Benoit Steiner2016-03-05
|
* Use the CMAKE_CXX_STANDARD variable to turn on cxx11Gravatar Benoit Steiner2016-03-04
|
* Don't rely on the M_PI constant since not all compilers provide it.Gravatar Benoit Steiner2016-03-04
|
* Fixed the computation of leading zeros when compiling with msvc.Gravatar Benoit Steiner2016-03-04
|
* MSVC uses __uint128 while other compilers use __uint128_t to encode 128bit ↵Gravatar Benoit Steiner2016-03-04
| | | | unsigned integers. Make the cxx11_tensor_uint128.cpp test work in both cases.
* Fixed syntax errorGravatar Benoit Steiner2016-03-04
|
* Added missing includeGravatar Benoit Steiner2016-03-04
|
* Don't use implicit type conversions in initializer lists since not all ↵Gravatar Benoit Steiner2016-03-04
| | | | compilers support them.
* Made the contraction test more portableGravatar Benoit Steiner2016-03-04
|
* Fixed a typoGravatar Benoit Steiner2016-03-04
|
* Added tests to cover the new rounding, flooring and ceiling tensor operations.Gravatar Benoit Steiner2016-03-03
|
* Added support for rounding, flooring, and ceiling to the tensor apiGravatar Benoit Steiner2016-03-03
|
* Added a test to validate the conversion of half floats into floats on Kepler ↵Gravatar Benoit Steiner2016-03-03
| | | | | | GPUs. Restricted the testing of the random number generation code to GPU architecture greater than or equal to 3.5.
* Enable partial support for half floats on Kepler GPUs.Gravatar Benoit Steiner2016-03-03
|
* Enable the conversion between floats and half floats on older GPUs that ↵Gravatar Benoit Steiner2016-03-03
| | | | support it.
* Merged in ebrevdo/eigen (pull request PR-167)Gravatar Benoit Steiner2016-03-03
|\ | | | | | | | | | | Add infinity() support to numext::numeric_limits, use it in lgamma. I tested the code on my gtx-titan-black gpu, and it appears to work as expected.
| * Small bugfix to numeric_limits for CUDA.Gravatar Eugene Brevdo2016-03-02
| |