Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Add EIGEN_HAS_INTRINSIC_INT128 macro | Rasmus Munk Larsen | 2019-11-06 |
| | | | | Add a new EIGEN_HAS_INTRINSIC_INT128 macro, and use this instead of __SIZEOF_INT128__. This fixes related issues with TensorIntDiv.h when building with Clang for Windows, where support for 128-bit integer arithmetic is advertised but broken in practice. | ||
* | [SYCL] This PR adds the minimum modifications to the Eigen unsupported ↵ | Mehdi Goli | 2019-06-28 |
| | | | | | | | | | | module required to run it on devices supporting SYCL. * Abstracting the pointer type so that both SYCL memory and pointer can be captured. * Converting SYCL virtual pointer to SYCL device memory in Eigen evaluator class. * Binding SYCL placeholder accessor to command group handler by using bind method in Eigen evaluator node. * Adding SYCL macro for controlling loop unrolling. * Modifying the TensorDeviceSycl.h and SYCL executor method to adopt the above changes. | ||
* | updates based on PR feedback | Deven Desai | 2018-06-14 |
| | | | | | | | | | | | | | | | | | There are two major changes (and a few minor ones which are not listed here...see PR discussion for details) 1. Eigen::half implementations for HIP and CUDA have been merged. This means that - `CUDA/Half.h` and `HIP/hcc/Half.h` got merged to a new file `GPU/Half.h` - `CUDA/PacketMathHalf.h` and `HIP/hcc/PacketMathHalf.h` got merged to a new file `GPU/PacketMathHalf.h` - `CUDA/TypeCasting.h` and `HIP/hcc/TypeCasting.h` got merged to a new file `GPU/TypeCasting.h` After this change the `HIP/hcc` directory only contains one file `math_constants.h`. That will go away too once that file becomes a part of the HIP install. 2. new macros EIGEN_GPUCC, EIGEN_GPU_COMPILE_PHASE and EIGEN_HAS_GPU_FP16 have been added and the code has been updated to use them where appropriate. - `EIGEN_GPUCC` is the same as `(EIGEN_CUDACC || EIGEN_HIPCC)` - `EIGEN_GPU_DEVICE_COMPILE` is the same as `(EIGEN_CUDA_ARCH || EIGEN_HIP_DEVICE_COMPILE)` - `EIGEN_HAS_GPU_FP16` is the same as `(EIGEN_HAS_CUDA_FP16 or EIGEN_HAS_HIP_FP16)` | ||
* | Fix typos found using codespell | Gael Guennebaud | 2018-06-07 |
| | |||
* | Hyperlink DOIs against preferred resolver | Katrin Leinweber | 2018-05-24 |
| | |||
* | Add a EIGEN_NO_CUDA option, and introduce EIGEN_CUDACC and EIGEN_CUDA_ARCH ↵ | Gael Guennebaud | 2017-07-17 |
| | | | | aliases | ||
* | Adding Tensor ReverseOp; TensorStriding; TensorConversionOp; Modifying ↵ | Mehdi Goli | 2017-01-16 |
| | | | | Tensor Contractsycl to be located in any place in the expression tree. | ||
* | Adding sycl backend for TensorPadding.h; disbaling __unit128 for sycl in ↵ | Mehdi Goli | 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. | ||
* | Fixing LLVM error on TensorMorphingSycl.h on GPU; fixing int64_t crash for ↵ | Mehdi Goli | 2016-11-25 |
| | | | | tensor_broadcast_sycl on GPU; adding get_sycl_supported_devices() on syclDevice.h. | ||
* | Emulate _BitScanReverse64 for 32 bits builds | Gael Guennebaud | 2016-07-11 |
| | |||
* | Change runtime to compile-time conditional. | Gael Guennebaud | 2016-07-08 |
| | |||
* | Fixed the integer division code on windows | Benoit Steiner | 2016-03-09 |
| | |||
* | Fixed the computation of leading zeros when compiling with msvc. | Benoit Steiner | 2016-03-04 |
| | |||
* | Fixed a typo | Benoit Steiner | 2016-03-04 |
| | |||
* | Updated the TensorIntDivisor code to work properly on LLP64 systems | Benoit Steiner | 2016-02-08 |
| | |||
* | Fixed the implementation of Eigen::internal::count_leading_zeros for MSVC. | Benoit Steiner | 2015-11-23 |
| | | | | Also updated the code to silence bogux warnings generated by nvcc when compilining this function. | ||
* | Added proper support for fast 64bit integer division on CUDA | Benoit Steiner | 2015-11-20 |
| | |||
* | Avoid using the version of TensorIntDiv optimized for 32-bit integers when ↵ | Benoit Steiner | 2015-11-18 |
| | | | | the divisor can be equal to one since it isn't supported. | ||
* | Fix some trivial warnings | Christoph Hertzberg | 2015-08-19 |
| | |||
* | Fixed 2 compilation warnings generated by llvm | Benoit Steiner | 2015-07-29 |
| | |||
* | Fixed a few compilation warnings triggered by clang | Benoit Steiner | 2015-07-29 |
| | |||
* | Simplified and generalized the DividerTraits code | Benoit Steiner | 2015-07-29 |
| | |||
* | Add missing specialization of struct DividerTraits<long> | Gael Guennebaud | 2015-07-29 |
| | |||
* | Extended the range of value inputs for TensorIntDiv to support tensors with ↵ | Benoit Steiner | 2015-07-22 |
| | | | | more than 4 billion elements. | ||
* | Fixed a bug in the integer division code that caused some large numerators ↵ | Benoit Steiner | 2015-07-13 |
| | | | | to be incorrectly handled | ||
* | Improved and cleaned up the 2d patch extraction code | Benoit Steiner | 2015-07-07 |
| | |||
* | Fix undefined behavior. | vanhoucke | 2015-06-19 |
| | |||
* | Added new version of the TensorIntDiv class optimized for 32 bit signed ↵ | Benoit Steiner | 2015-05-19 |
| | | | | integers. It saves 1 register on CPU and 2 on GPU. | ||
* | Fixed incorrect assertion | Benoit Steiner | 2015-02-28 |
| | |||
* | Fixed another batch of compilation warnings | Benoit Steiner | 2015-02-28 |
| | |||
* | Fixed another compilation problem with TensorIntDiv.h | Benoit Steiner | 2015-02-26 |
| | |||
* | Made TensorIntDiv.h compile with MSVC | Benoit Steiner | 2015-02-25 |
| | |||
* | Fixed another clang warning | Benoit Steiner | 2015-02-25 |
| | |||
* | Misc improvements and cleanups | Benoit Steiner | 2014-10-13 |
| | |||
* | Added support for fast integer divisions by a constant | Benoit Steiner | 2014-08-14 |
Sped up tensor slicing by a factor of 3 by using these fast integer divisions. |