Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | asm escape was not strong enough to prevent too aggressive compiler ↵ | 2016-07-04 | |
| | | | | optimization let's fallback to no-inline. | ||
* | Few fixes in perf-monitoring. | 2016-07-04 | |
| | |||
* | fix clang compilation | 2016-07-04 | |
| | |||
* | Workaround compilation issue with msvc | 2016-07-04 | |
| | |||
* | Fix compilation of some unit tests with msvc | 2016-07-04 | |
| | |||
* | Change the semantic of the last template parameter of Assignment from ↵ | 2016-07-04 | |
| | | | | | | | "Scalar" to "SFINAE" only. The previous "Scalar" semantic was obsolete since we allow for different scalar types in the source and destination expressions. On can still specialize on scalar types through SFINAE and/or assignment functor. | ||
* | Fix performance regression in dgemm introduced by changeset ↵ | 2016-07-02 | |
| | | | | 5d51a7f12c69138ed2a43df240bdf27a5313f7ce | ||
* | Fix performance regression introduced in changeset ↵ | 2016-07-02 | |
| | | | | | | | | | e56aabf205a1e8f581dd8a46d7d46ce79c45e158 . Register blocking sizes are better handled by the cache size heuristics. The current code introduced very small blocks, for instance for 9x9 matrix, thus killing performance. | ||
* | Merged in carpent/eigen (pull request PR-204) | 2016-07-01 | |
|\ | | | | | | | Use complete nested namespace Eigen::internal, thus making the custom static assertion macros available outside the Eigen's namespace. | ||
* | | Made it possible to compile reductions for an old cuda architecture and run ↵ | 2016-06-29 | |
| | | | | | | | | them on a recent gpu. | ||
* | | Made the code compile when using CUDA architecture < 300 | 2016-06-29 | |
| | | |||
* | | Merged in ibab/eigen/fix-tensor-scan-gpu (pull request PR-205) | 2016-06-29 | |
|\ \ | | | | | | | | | | Add missing CUDA kernel to tensor scan op | ||
| * | | Add missing CUDA kernel to tensor scan op | 2016-06-29 | |
|/ / | | | | | | | | | The TensorScanOp implementation was missing a CUDA kernel launch. This adds a simple placeholder implementation. | ||
| * | Use complete nested namespace Eigen::internal | 2016-06-28 | |
|/ | |||
* | Undo changes in AltiVec --- I don't have any way to test there. | 2016-06-28 | |
| | |||
* | Avoid global variables with static constructors in NEON/Complex.h | 2016-06-28 | |
| | |||
* | Added a test to validate the tensor scan evaluation on GPU. The test is ↵ | 2016-06-27 | |
| | | | | currently disabled since the code segfaults. | ||
* | Don't store the scan axis in the evaluator of the tensor scan operation ↵ | 2016-06-27 | |
| | | | | | | since it's only used in the constructor. Also avoid taking references to values that may becomes stale after a copy construction. | ||
* | bug #1247: fix regression in compilation of pow(integer,integer), and add ↵ | 2016-06-25 | |
| | | | | respective unit tests. | ||
* | Fix hyperbolic functions for autodiff. | 2016-06-24 | |
| | |||
* | Fix missing specialization. | 2016-06-24 | |
| | |||
* | merge pull request 198 | 2016-06-24 | |
|\ | |||
| * | Fix some unused typedef warnings. | 2016-06-24 | |
| | | |||
| * | Fix NumTraits<AutoDiff> | 2016-06-24 | |
| | | |||
| * | Fix instantiation of ScalarBinaryOpTraits for AutoDiff. | 2016-06-24 | |
| | | |||
| * | Relax promote_scalar_arg logic to enable promotion to Expr::Scalar if ↵ | 2016-06-24 | |
| | | | | | | | | | | | | | | conversion to Expr::Literal fails. This is useful to cancel expression template at the scalar level, e.g. with AutoDiff<AutoDiff<>>. This patch also defers calls to NumTraits in cases for which types are not directly compatible. | ||
| * | bug #1245: fix compilation with msvc | 2016-06-24 | |
| | | |||
| * | Return -1 from CurrentThreadId when called by thread outside the pool. | 2016-06-23 | |
| | | |||
| * | Resolve merge. | 2016-06-23 | |
| |\ | |||
| | * | Add unit test for printing empty tensors | 2016-06-23 | |
| | | | |||
| | * | bug #1241: does not emmit anything for empty tensors | 2016-06-23 | |
| | | | |||
| | * | Fix operator priority | 2016-06-23 | |
| | | | |||
| | * | Fix warning. | 2016-06-23 | |
| | | | |||
| | * | merge PR 194 | 2016-06-23 | |
| | |\ | |||
| | | * | Introduce a NumTraits<T>::Literal type to be used for literals, and | 2016-06-23 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | improve mixing type support in operations between arrays and scalars: - 2 * ArrayXcf is now optimized in the sense that the integer 2 is properly promoted to a float instead of a complex<float> (fix a regression) - 2.1 * ArrayXi is now forbiden (previously, 2.1 was converted to 2) - This mechanism should be applicable to any custom scalar type, assuming NumTraits<T>::Literal is properly defined (it defaults to T) | ||
| | * | | Biug 1242: fix comma init with empty matrices. | 2016-06-23 | |
| | | | | |||
| | * | | Silenced a couple of compilation warnings generated by xcode | 2016-06-22 | |
| | | | | |||
| | * | | Turned the constructor of the PerThread struct into what is effectively a ↵ | 2016-06-22 | |
| | | | | | | | | | | | | | | | | constant expression to make the code compatible with a wider range of compilers | ||
| | * | | Handle empty tensors in the print functions | 2016-06-21 | |
| | | | | |||
| | * | | Fixed the printing of rank-0 tensors | 2016-06-20 | |
| | | | | |||
| | * | | Made log1p_impl usable inside a GPU kernel | 2016-06-16 | |
| | | | | |||
* | | | | Add autodiff coverage for standard library hyperbolic functions, and tests. | 2016-06-15 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | * * * Corrected tanh derivatived, moved test definitions. * * * Added more test cases, removed lingering lines | ||
| | | * | Fix warnings with gcc | 2016-06-15 | |
| | | | | |||
| | | * | Move scalar/expr to ArrayBase and fix documentation | 2016-06-15 | |
| | | | | |||
| | | * | Propagate functor to ScalarBinaryOpTraits | 2016-06-15 | |
| | | | | |||
| | | * | Include the cost of stores in unrolling of triangular expressions. | 2016-06-15 | |
| | | | | |||
| | * | | Merged in ibab/eigen (pull request PR-197) | 2016-06-14 | |
| | |\ \ | | | | | | | | | | | | | | | | Implement exclusive scan option for Tensor library | ||
| | * | | | Avoid generating pseudo random numbers that are multiple of 5: this helps | 2016-06-14 | |
| | | | | | | | | | | | | | | | | | | | | spread the load over multiple cpus without havind to rely on work stealing. | ||
| | | | * | Cleanup useless helper: internal::product_result_scalar | 2016-06-15 | |
| | | | | | |||
| | | | * | Include the cost of stores in unrolling (also fix infinite unrolling with ↵ | 2016-06-15 | |
| | | | | | | | | | | | | | | | | | | | | expression costing 0 like Constant) |